1) Could you explain the significance of your article to the non-specialist?

One of the mysteries of research on glassy water is whether it turns on heating at atmospheric pressure between ~140-150 K into an ultraviscous liquid, or whether it remains a glass up to its crystallization to cubic ice. We now show by differential scanning calorimetry of hyperquenched water deposited at 140 K and subjected to various cooling-heating cycles and annealing treatment that its relaxation behaviour at these low temperatures is that of a liquid. The results are consistent with the theory of glass-to-liquid transition and thus, water's heat capacity increase at ~136 K indicates the onset of a glass-to-liquid transition, and not of a so-called "shadow peak".

 

2) What has motivated you to conduct this work?

It has been vigorously debated whether the glass transition of hyperquenched water occurs at ~136 K, or whether the endothermic feature observed by calorimetry is just a "shadow" of a glass transition at ~165 K. Our motivation was, therefore, to clarify this issue by investigating for the first time the effect of various cooling rates and annealing procedures on water's heat capacity increase at ~136 K.

 

3) Where do you see this work developing in the future?

Our new data show that between ~140-150 K glassy water turns into an ultraviscous liquid. New measurements of its thermodynamic and transport properties are now possible and meaningful in order to clarify the nature of liquid water's ultraviscous and deeply supercooled state, and to connect its properties with those at ambient and supercooled temperatures. This should lead to an understanding of the anomalies of supercooled water. Furthermore, our new data support the proposal of a fragile-to-strong transition on cooling liquid water from ambient temperature into the deeply supercooled and glassy state. They show that the previously reported value of the heat capacity increase at ~136 K must contain a contribution from an overshoot caused by the annealing procedure. It follows that the "true" value is even lower and that this could classify deeply supercooled water as an even stronger liquid than SiO2. This has certainly implications on our understanding of water's hydrogen-bonded network and of its structural changes with temperature.

 

4) Are there any particular challenges facing future research in this area?

The main challenge is that glassy water crystallizes on heating at 1 bar at ~150 K, and thus, that only a small temperature region between ~140-150 K is open for experimentation on its ultraviscous liquid state. The temperature region between ~150-230 K is not accessible for experimentation because formation of ice occurs too fast, and because of that it has been called "no-man´s land". Even in the small window region between ~140-150 K formation of ice is possible, and thus experimental measurements operating on short time scales are needed in order to avoid interference by crystallization.