Effects of hydrogen bonding on supercooled liquid dynamics and the implications for supercooled water

TL;DR

This study uses hydrogen-bonded oligomeric liquids to model supercooled water, revealing a dynamic transition around 220 K and suggesting water's behavior is similar to other supercooled liquids, despite experimental inaccessibility.

Contribution

It introduces a novel approach of studying oligomeric glycols to infer supercooled water dynamics, predicting a dynamic transition at 220 K.

Findings

Dynamic crossover observed in glycols' relaxation times.

Predicted dynamic transition in water at ~220 K.

Evidence for secondary relaxation in supercooled water.

Abstract

The supercooled state of bulk water is largely hidden by unavoidable crystallization, which creates an experimentally inaccessible temperature regime - a 'no man's land'. We address this and circumvent the crystallization problem by systematically studying the supercooled dynamics of hydrogen bonded oligomeric liquids (glycols), where water corresponds to the chain-ends alone. This novel approach permits a 'dilution of water' by altering the hydrogen bond concentration via variations in chain length. We observe a dynamic crossover in the temperature dependence of the structural relaxation time for all glycols, consistent with the common behavior of most supercooled liquids. We find that the crossover becomes more pronounced for increasing hydrogen bond concentrations, which leads to the prediction of a marked dynamic transition for water within 'no man's land' at T~220 K. Interestingly, the predicted transition thus takes place at a temperature where a so called 'strong-fragile' transition has previously been suggested. Our results, however, imply that the dynamic transition of supercooled water is analogous to that commonly observed in supercooled liquids. Moreover, we find support also for the existence of a secondary relaxation of water with behavior analogous to that of the secondary relaxation observed for the glycols.