Compression-rate dependence of the phase transition from hexagonal ice to ice II and/or ice III

We study the density-driven phase transition from hexagonal ice in the temperature range of 170–230 K up to pressures of 0.65 GPa using a piston-cylinder apparatus. Pure ice II, pure ice III, or mixtures of ice II and ice III are identified as products when the compression-rate is varied from 0.001–4 GPa/min. At low compression-rates and at high temperatures, formation of ice II is observed, which is in accordance with the phase diagram. However, at low temperatures and at higher compression rates, formation of metastable ice III is observed, which extends the known temperature range for possible formation of ice III. Metastable ice III rather than stable ice II crystals are produced by simple variation of the compression rate, which is an uncommon concept of producing metastable phases. We discuss some implications for our understanding of the interior of icy satellites such as Ganymede.

Figure 1

(Color online) Water’s phase diagram of stable phases. The shaded area indicates the approximate pressure and temperature conditions in the interior of the icy satellites of outer planets. The arrows indicate the paths taken for the uniaxial compression experiments of hexagonal ice. The figure was adapted from Leliwa-Kopystynski (Ref. 10).

Figure 2

(Color online) (a) Piston displacement vs nominal pressure curves showing the densification step to ice II at a rate of 10 MPa/min (black solid line) and to ice III at a rate of 1000 MPa/min (red dashed line). The onset of the phase transition is indicated by black dotted tangents. In addition the temperature as read out by the Pt-100 sensor mounted in the steel cylinder 10 mm from the sample is shown (right axis). (b) Powder x-ray diffractograms of the samples shown in (a) after quenching to 77 K and decompression to 1 bar as measured in $\theta \text{−}\theta$ geometry at $\sim 80\text{K}$ using $\text{Cu-}K\alpha$ radiation. Ticks close to the powder x-ray diffractograms indicate the expected peak position from the structures of ice II (Refs. 18, 20) and ice III (Ref. 21). (c) Piston displacement vs nominal pressure curves showing the densification step to ice III at 170 K (red dashed line) to a 85:15 mixture of ice III and ice II at 198 K (blue dotted line) and to ice II at 230 K (black solid line), all at a compression rate of 100 MPa/min. The onset of the phase transition is indicated by black dotted tangents. (d) Powder x-ray diffractograms of the samples shown in (c) after quenching to 77 K and decompression to 1 bar as measured in $\theta \text{−}\theta$ geometry at $\sim 80\text{K}$ using $\text{Cu-}K\alpha$ radiation. Ticks close to the powder x-ray diffractograms indicate the expected peak position from the structures of ice II (Refs. 18, 20) and ice III (Ref. 21).

Figure 3

The effect of variation of compression rate on the fraction of ice III in the ice II/ice III mixture produced by compressing hexagonal ice isothermally at 198 K. The error bars indicate the reliability of the POWDER CELL (Ref. 44) fit of ice II and ice III patterns to the observed powder x-ray diffractograms.