We report relaxation dynamics of glycerol-water mixtures as probed by megahertz-to-terahertz dielectric spectroscopy in a frequency range from 50 MHz to 0.5 THz at room temperature. The dielectric relaxation spectra reveal several polarization processes at the molecular level with different time constants and dielectric strengths, providing an understanding of the hydrogen-bonding network in glycerol-water mixtures. We have determined the structure of hydration shells around glycerol molecules and the dynamics of bound water as a function of glycerol concentration in solutions using the Debye relaxation model. The experimental results show the existence of a critical glycerol concentration of ~7.5 mol %, which is related to the number of water molecules in the hydration layer around a glycerol molecule. At higher glycerol concentrations, water molecules dispersed in a glycerol network become abundant and eventually dominate and four distinct relaxation processes emerge in the mixtures. The relaxation dynamics and hydration structure in glycerol-water mixtures are further probed with molecular dynamics simulations, which confirm the physical picture revealed by the dielectric spectroscopy.

Glass transition temperature (Tg) plays an important role in controlling the mechanical and thermal properties of a polymer. Polyimides are an important category of polymers with wide applications because of their superior heat resistance and mechanical strength. The capability of predicting Tg for a polyimide a priori is therefore highly desirable in order to expedite the design and discovery of new polyimide polymers with targeted properties and applications. Here we explore three different approaches to either compute Tg for a polyimide via all-atom molecular dynamics (MD) simulations or predict Tg via a mathematical model generated by using machine-learning algorithms to analyze existing data collected from literature. Our simulations reveal that Tg can be determined from examining the diffusion coefficient of simple gas molecules in a polyimide as a function of temperature and the results are comparable to those derived from data on polymer density versus temperature and actually closer to the available experimental data. Furthermore, the predictive model of Tg derived with machine-learning algorithms can be used to estimate Tg successfully within an uncertainty of about 20 degrees, even for polyimides yet to be synthesized experimentally.

We use molecular dynamics simulations to probe hydration, ion spacing, and cation–anion interaction in two sulfonated polysulfones with different ion distributions along the polymer backbone. At room temperature, these polymers are below their experimental glass transition temperatures even with water contents more than 10%. At the equilibrium water uptake, the ions exhibit a similar level of hydration as they would in their saturated aqueous solution. The framework of Manning’s limiting law for counterion condensation is used to examine ionic interactions in the simulated polysulfones. The dielectric constant (ε) that the ions experience can be well approximated by a volume-weighted average of the dielectric constants of the polymer backbone and water. Our results show that a reasonable estimate of the average interionic distance, b, is obtained by using the distance where the sulfonate–sulfonate coordination number reaches 1. The spacing of the sulfonate ions along the polysulfone backbone plays a role in determining their spatial distribution inside the hydrated polymer. As a result, the value of b is slightly larger for polymers where the sulfonate ions are more evenly spaced along the backbone, which is consistent with experimental evidence. The simulations reveal that the sulfonate ions and sodium counterions form fibrillar aggregates at water contents below the equilibrium water uptake. Such extensive ionic aggregates are expected to facilitate ion transport in sulfonated polysulfone membranes without the need for long-range chain motion as in the case of traditional rubbery ionic polymers. Our estimates for ε and b are used in conjunction with Manning’s theory to estimate the fraction of counterions condensed to the fixed ions. The prediction of Manning’s theory agrees well with the result computed by directly counting the condensed sodium ions in the molecular dynamics simulations.

We investigate the behavior of hydrated sulfonated polysulfones over a range of ion contents through differential scanning calorimetry (DSC), Fourier transform infrared (FTIR) spectroscopy, and molecular dynamics (MD) simulations. Experimental evidence shows that at comparable ion contents, the spacing between the ionic groups along the polymer backbone can significantly impact the amount of melting water present in the polymer. When we only consider water molecules that can hydrogen-bond to four neighboring water molecules as the melting water, the MD simulation results are found to agree with the experimental data. The states of water measured by DSC can therefore be described as "aggregated" (or bulk-like) for the melting component and "isolated" for the nonmelting part. Using this physical picture, a polymer with more aggregated ions has a higher content of melting water, while a polymer at the same ion content but with more dispersed ions has a lower content of melting water. Therefore, ions should be well dispersed to minimize the amount of bulk-like water in ionic polymer membranes.

No abstract