The dielectric response of ice explained

Japanese researchers have explained a decades-old discrepancy concerning the dielectric response of ice. Publishing their findings in The Journal of Physical Chemistry B, the scientists show that previous differences in study results likely arise as a result of the ice preparation conditions and resulting impurity levels.

A dielectric material is an electrical insulator that can be polarised by an electric field. In 1952, Auty and Cole published results from studies of water ice’s dielectric relaxation time, τ_ice — the lag in the response of molecules to changes in the electric field. They found a linear ‘Arrhenius’ relationship between τ_ice and temperature, suggesting that reaction rates double with every 10 K increase in temperature between 207 and 273 K.

In 1981, however, Johari and Whalley reported observed changes in the relationship at 230 and 140 K. Subsequent experiments using various parameters — thereby excluding possible explanations based on crystal size or a metastable crystal surface — have also noted temperatures at which the dielectric behaviour apparently crosses over to a new regime.

Kaito Sasaki, Rio Kita, Naoki Shinyashiki and Shin Yagihara from Tokai University prepared ice in three different ways and found that agreement with Auty et al or Johari et al depended on the preparation process. Arrhenius relationships were observed in ice for all three preparation techniques, although below a certain temperature they cross over into another temperature-dependent dielectric regime. For ice slow-cooled with a magnetic stirrer, this crossover was not observed down to temperatures below that reported by Auty et al. The researchers suggest this is due to the absence of impurity-produced orientational defects.

The conundrum also affects studies of proteins and glycerol dissolved in ice. In all studies of the dielectric properties of proteins dissolved in aqueous ice, an unknown relaxation process is observed within a similar frequency range. Understanding the dielectric properties of ice can help to understand the contributions to the dielectric response of protein solutions and their viscoelastic properties.

“An explanation of the mechanism underlying the difference between the τ_ice (Auty-1952) and τ_ice (Johari-1981) would enable the use of τ_ice observed in aqueous systems to be utilised effectively not only for understanding ice dynamics but also as a probe of the dynamic viscoelastic behaviour of water and solute molecules through ice growth,” the study authors wrote.