Reply to “Comment on ‘Spontaneous liquid-liquid phase separation of water’ ”

Two different scenarios have been proposed on the phase separation occurring in the deeply supercooled liquid water. We discuss what we can derive from our simulation results for the two scenarios and propose a way for future investigation. We also demonstrate that the phase separation in the supercooled liquid water looks like the separation of liquid water and vapor just below the conventional critical point.

Figure 1

Density profiles of water near the vapor-liquid critical point of TIP4P/2005 water. Panels (a)–(c) are the density profiles at $T=650,625,\mathrm{and}610\mathrm{K}$, respectively. $NVT$ simulations were performed at $0.31\mathrm{g}\mathrm{c}{\mathrm{m}}^{-3}$. The time axis is scaled by the relaxation time of the hydrogen bond correlation function. The relaxation times at 650, 625, and 610 K are 226, 258, and 280 fs, respectively. The reduced temperatures $T_{\mathrm{r}}=T/T_{\mathrm{c}}$ are 1.013, 0.974, and 0.951.

Figure 2

Density profiles of supercooled ST2 water at (a) 250 K, (b) 240 K, and (c) 235 K. The trajectories are the same as those presented in Fig. 1 of Ref. [1]. The time axis is scaled by the relaxation time of the hydrogen bond correlation function. The relaxation times at 250, 240, and 235 K are 84, 311, and 1180 ps, respectively. The reduced temperatures $T_{\mathrm{r}}$ are 1.012, 0.972, and 0.951 ($T_{\mathrm{c}}=247\mathrm{K}$ is taken from Ref. [14]).