Molecular Dynamics Simulation of the Influence of External Electric Fields on the Glass Transition Temperature of the Ionic Liquid 1-Ethyl-3-methylimidazolium Bis(trifluoromethylsulfonyl)imide

Abstract

We present the results of molecular dynamics simulations of the ionic liquid (IL) 1-ethyl-3-methylimidazolium bis­(trifluoromethylsulfonyl)­imide [C2C1im]­[NTf2] in the presence of external electric fields (EEFs) of varying strengths to understand the effects of EEFs on the glass transition temperature T g. We compute T g with an automated and objective method and observe a depression in T g when cooling the IL within an EEF above a critical strength. The effect is reversible, and glasses prepared with EEFs recover their original zero-field T g when heated. By examining the dynamics and structure of the liquid phase, we find that the EEF lowers the activation energy for diffusion, reducing the energetic barrier for movement and consequently T g. We show that the effect can be leveraged to drive an electrified nonvapor compression refrigeration cycle.

Glass Transition and Structure of Organic Polymers from All-Atom Molecular Simulations

“…This approach will be termed the minimum residual method hereafter. Similarly to the intersection method, it assumes that the observed quantity for glass and liquid phases behaves linearly with distinct slopes, but it does not omit any data points close to the expected glass transition . This minimum residual analysis of nonequilibrium data eliminates some degrees of uncertainty and arbitrariness introduced in the intersection approach.…”

“…Computational straightforwardness and low uncertainty of bulk densities render this property optimal for the evaluation of the glass transition temperature by the trend shift method . Typically, a bilinear fit interpolating the dependences of the density of liquid and glass on temperature separately can be employed. , Such an approach contained some arbitrariness in the assignment of the temperature intervals of both regimes to be interpolated linearly so that automated protocols have been devised. , These exploit either a more sophisticated hyperbola fit to circumvent the deviations from linear trends close to the actual glass transition temperature , or a protocol minimizing the residual squares related to concurrent bilinear interpolation of the liquid and glassy branches without omitting any data points close to the expected trend shift. , …”

“…14,18 These exploit either a more sophisticated hyperbola fit to circumvent the deviations from linear trends close to the actual glass transition temperature 19,20 or a protocol minimizing the residual squares related to concurrent bilinear interpolation of the liquid and glassy branches without omitting any data points close to the expected trend shift. 14,21 Underlying density versus temperature data can be obtained from separate simulations for each individual considered temperature, targeting to reach equilibrium within each simulation. Such an approach is relatively costly; it may suffer from the fact that the amorphous solid state is not related to any equilibrium, and its simulations may freeze in coincidental states at low temperatures.…”

“…An alternative nonequilibrium approach consists of performing a gradual cooling of a molten system down to a frozen state. 21 While this approach is computationally more efficient, imposing too fast cooling rates does not guarantee that adequate response of the density to the temperature variation is captured within a limited trajectory. It is desirable to repeat the cooling runs from independent initial molten states to avoid freezing in a coincidental state of low relevance and to glimpse a true computational uncertainty.…”

“…It is desirable to repeat the cooling runs from independent initial molten states to avoid freezing in a coincidental state of low relevance and to glimpse a true computational uncertainty. 21 In general, any molecular-dynamics-based predictions of the glass transition temperatures are prone to imposing too fast cooling rates (≈10 9 K s −1 ) that are orders of magnitude far from the conditions relevant for real experiments (≈10 −2 to 10 −1 K s −1 ), 22 which can affect the calculated results.…”