Impact of $\mathrm{CaC}{\mathrm{l}}_2$-induced chemical pressure on the phase transition of ${\mathrm{H}}_2\mathrm{O}$ at high pressure

Understanding the phase behavior and structural properties of salt water at high pressures is essential for understanding the dynamics and physical characteristics of icy planets. In this study, we employed high-pressure experimental and ab initio simulation techniques to investigate the impact of $\mathrm{CaC}{\mathrm{l}}_2$ on the structure of ice VII. Our findings reveal that 1.8 mol% $\mathrm{CaC}{\mathrm{l}}_2$ can be incorporated into the ice VII structure above 10 GPa. This $\mathrm{CaC}{\mathrm{l}}_2$-bearing ice VII (Cb VII) exhibits a lower O-H stretching frequency in the Raman spectra as well as a reduced volume of the unit cell compared to pure ice VII. In contrast to doping ice VII with other salts such as LiCl and NaCl that leads to an increase of the ice VII to ice X transition pressure occurring at 100–150 GPa, $\mathrm{CaC}{\mathrm{l}}_2$ doping stands out by reducing the transition pressure. It shifts the transition to a pressure of 52 GPa, which is significantly lower than the transition pressure of 80 GPa in the pure ${\mathrm{H}}_2\mathrm{O}$ ice system. This notable distinction highlights the unique influence of $\mathrm{CaC}{\mathrm{l}}_2$ on the phase behavior of water under high pressure, and we attribute these effects to the phenomenon of chemical pressure induced by $\mathrm{CaC}{\mathrm{l}}_2$ within the ice VII structure. Our study suggests that the presence of a modified ice VII phase, contaminated with salt and referred to as Cb VII, may influence the composition, structure, and evolution of planets.

Figure 1

(a) Representative Raman spectra of ice VI, ice VII, and $\mathrm{CaC}{\mathrm{l}}_2$ hydrate from the 1.8 mol% $\mathrm{CaC}{\mathrm{l}}_2$ solutions with increasing pressure at 298 K. The Raman modes of liquid, ice VI, ice VII, and $\mathrm{CaC}{\mathrm{l}}_2$ hydrate are indicated by yellow, green, and blue solid squares, and red solid stars, respectively. (b) XRD patterns of 1.8 mol% $\mathrm{CaC}{\mathrm{l}}_2$ solutions at high pressures and 298 K. The sequence of patterns represents the evolution during the compression experiment from 2.6 to 25 GPa. Colored tick marks show the diffraction peak positions of gold (green), ice VII (blue), and $\mathrm{CaC}{\mathrm{l}}_2$ hydrate (red). The patterns relative to undissociated $\mathrm{CaC}{\mathrm{l}}_2$ hydrate are plotted in black, while those of the dissociated $\mathrm{CaC}{\mathrm{l}}_2$ hydrate are plotted in blue.

Figure 2

(a) O-H stretching region of the Raman spectrum of pure ice VII and $\beta \text{−}\mathrm{CaC}{\mathrm{l}}_2·4{\mathrm{H}}_2\mathrm{O}$. The cyan solid line shows the Lorentzian peak fit for signals of ice VII. $A_{1\mathrm{g}}, E_{1\mathrm{g}}$, and $B_{1\mathrm{g}}$ peaks of ice VII correspond to the pink, green, and blue solid lines, respectively. The orange solid line shows the double Lorentzian peak fit for signals of ice VII and $\beta \text{−}\mathrm{CaC}{\mathrm{l}}_2·4{\mathrm{H}}_2\mathrm{O}$. The vibration modes 1 and 2 of $\beta \text{−}\mathrm{CaC}{\mathrm{l}}_2·4{\mathrm{H}}_2\mathrm{O}$ are depicted in red and fluorescent green lines, respectively. The black and blue dotted lines represent our experimental results of pure ice VII and ice VII with $\beta \text{−}\mathrm{CaC}{\mathrm{l}}_2·4{\mathrm{H}}_2\mathrm{O}$. (b) O-H stretching region of the Raman spectrum of pure ice VII and Cb VII. The green dotted lines represent our experimental result of Cb VII. The gray solid line shows the double Lorentzian peak fit for signals of ice VII and Cb VII. The vibration modes 1, 2, and 3 of Cb VII are depicted by yellow, purple, and yellow lines, respectively.

Figure 3

(a) Evolution of $\mathrm{\Delta }{\nu }_{\mathrm{OH}} \left(\mathrm{c}{\mathrm{m}}^{-1}\right)$ as a function of pressure for ice VII formed from a 1.8 mol% $\mathrm{CaC}{\mathrm{l}}_2$ solution. Data from this study are depicted in red solid circles. The open circles represent the results of previous experiments [26]. The purple dashed line represents the pressure at which $\beta \text{−}\mathrm{CaC}{\mathrm{l}}_2·4{\mathrm{H}}_2\mathrm{O}$ undergoes dissociation. (b) P-V relations of $\mathrm{CaC}{\mathrm{l}}_2$-bearing ice VII, NaCl-bearing ice VII, and pure ice VII. Data from this study are depicted in red solid circles. The red solid line shows the fitting results using the third-order Birch-Murnaghan EoS in this study. The dashed lines represent the results of previous experiments [10, 26, 59, 60].

Figure 4

Illustrations of nine different $\mathrm{CaC}{\mathrm{l}}_2$ configurations in ice VII structure before ionic relaxation. Different color spheres in the sites represent H (pink), Ca (blue), Cl (green), and O (red) atoms, respectively.

Figure 5

The percentage change in volume from ab initio calculations between each water molecule in the nine configurations and each water molecule in ice VII under different pressures. The gray dashed line represents the volume of ice VII from DFT calculations. Different color solid circles represent the percentage increase in calculated volume of Cb VII 30-a (blue), 30-b (cyan), 30-c (pink), 62-a (yellow), 62-b (orange), 62-c (purple), 58-a (black), 58-b (red), and 56-c (green), respectively. The data of Cb VII 58-b overlaps with the data of Cb VII 58-a.

Figure 6

(a) Unit-cell volume of pure and salty ice at high pressures and 298 K. Red solid circles represent our experimental result. Green dashed circles: NaCl-bearing ice VII (Frank et al. [36]); purple dashed circles: NaCl-bearing ice VII (Shi et al. [65]); orange circles: pure ice (Hemley et al. [10]); pink circles: pure ice (Frank et al. [59]); cyan circles: pure ice (Sugimura et al. [16]). The pink shade indicates the pressure range for the dynamically disordered ice X′ estimated from XRD measurements. (b) Evolution of the ice $\mathrm{X}\text{−}{T}_{2\mathrm{g}}$ mode on compression above 47 GPa. The solid red line represents the EoS of this study. The green dashed line represents the EoS of pure ice VII.

Figure 7

Radial distribution function $g$($r$) of the distance OH below the transition (in phase VII at approximately 30 GPa) in pure ice VII (in red) [63] and in $\mathrm{CaC}{\mathrm{l}}_2\text{−}\mathrm{bearing}$ ice VII (in blue), obtained from quantum thermal bath molecular dynamics. The dotted lines indicate the average OH covalent bond length, while the dashed lines indicate the average O· · ·H hydrogen bond length.