A skillful experimenter can chill water to tens of degrees below 0 °C without having it freeze. If the liquid is free of impurities, it will persist in a metastable, supercooled state. Below about −40 °C, however, its fate is a matter of debate. The so-called stability-limit conjecture posits that the liquid phase destabilizes, and effectively ceases to exist, at about −45 °C. But other theories suggest that it can survive to far colder temperatures—and that a second liquid phase may even arise. The problem is vexing in part because the answer lies in an experimental no-man’s-land: Between roughly −40 °C and −125 °C, liquid water, assuming it exists, crystallizes seemingly too fast for an observer to confirm it was ever there.
Now Greg Kimmel, Bruce Kay, and their colleagues at Pacific Northwest National Laboratory have forged an experimental path into no-man’s-land. First, they deposited water onto a cryogenically cooled surface to make a thin film of amorphous ice—solid water frozen in a liquid-like molecular configuration. Then they irradiated the film with nanosecond IR pulses. Each pulse melts the film and warms it to no-man’s-land temperatures, but ever so briefly, so that the film crystallizes only partially before being quenched back to the amorphous ice state. Using surface-science techniques, the team could determine how much new ice formed with each pulse.
From the ice-formation rates, the researchers deduced that water’s diffusivity varies smoothly with temperature throughout no-man’s-land. That effectively rules out the stability-limit conjecture, which predicts a sharp kink. But it leaves unsettled the question of a second liquid phase. To test that theory, the researchers would have to adapt their vacuum-based technique for high-pressure operation, a task that Kay says “would be tricky, but possibly doable.” (Y. Xu et al., Proc. Natl. Acad. Sci. USA 113, 14921, 2016.)
