Does a liquid-liquid phase transition exist in supercooled water

Partly settled. Some of it is understood, some is still argued about.

In 1992, Poole, Sciortino, Essmann and Stanley proposed that water's many anomalies could be explained by a hidden liquid-liquid critical point deep in the supercooled, hard-to-access 'no man's land' region, based on computer simulations. Testing this experimentally was extraordinarily difficult because water crystallizes into ice on microsecond timescales in that regime, faster than most measurements could be made. In 2020, Kim and colleagues used ultrafast X-ray probes on femtosecond-laser-heated high-density amorphous ice to catch water in this transient liquid state, reporting density fluctuations consistent with a liquid-liquid transition before crystallization set in. The result was widely reported as strong evidence for the hypothesis, but debate continues in the literature over alternative interpretations of the data, so it is best characterized as a major but not fully settled advance rather than a closed question.

Water, the most familiar liquid on Earth, may secretly have two different liquid forms of itself hiding in a sliver of temperature and pressure so extreme it can only exist for microseconds.

Explanations that circulated

• Some researchers have argued the 2020 Science result could instead reflect artifacts such as ice nanocrystallite contamination rather than a genuine liquid-liquid phase transition; this critique and its rebuttals should be checked for the current state of consensus

References

• Experimental observation of the liquid-liquid transition in bulk supercooled water under pressure, Kyung Hwan Kim et al., Science, 2020