The dielectric loss ϵ′ in broadband dielectric spectroscopy (BDS) experiments as well as the imaginary part of the generalized susceptibility χ′ in depolarized dynamic light scattering (DDLS)experiments contains several peaks that can be assigned to different processes of molecular dynamics.Here, the spectral shape of the α-process is of particular interest, as it is linked to the macroscopic viscosity.

When comparing different glass-forming liquids in the BDS, the α-process shows strongly deviating shape parameters β_KWW of the corresponding correlation function, which typically lie between 0.5 and 0.8. In DDLS, a different picture emerges. Here a large number of liquids show the same “generic” spectral shape which at high frequencies can be described by a power law ∝ ω^−1/2. The figure shows data of various glass-forming liquids measured by BDS (left) and DDLS (right). The set of substances includes polar liquids with strongly different polarity, van der Waals liquids, hydrogen bonding systems and some ionic liquids.

In order to understand these discrepancies in spectral shape when using different methods, the methods must be considered in more detail. The experimentally determined correlation and response functions usually contain not only a self-correlation component but also contributions from cross-correlations which are due to interactions between molecules such as dipole-dipole interactions. BDS is particularly sensitive for the latter while DDLS provides more direct access to the self-correlation component. Thus, the deviation of the spectral shape from generic behavior in BDS could therefore possibly be explained by cross-correlations.

This simple fact already leads to drastic changes in phase transition temperatures in hydrogen bonding systems (for example, the melting temperature of alcohols is about 100° higher than that of comparable alkanes) and to the formation of – sometimes temporary – supramolecular structures. In particular the latter property has far-reaching consequences for a variety of physical, chemical and biological processes. Starting with monohydroxy alcohols (only one OH group), the complexity increases with the number of H-bonds.

In order to better understand the individual relaxation mechanisms, the entire range of experimental techniques is used. It is useful to supercool the sample under investigation, since the respective processes can be dynamically separated and thus better assigned to their origin.

Even the supposedly simple monohydroxy alcohols show several dynamic processes whose origin is not yet fully understood. Especially the so-called Debye process has been of major interest for along time. It is being associated with the reorientation of transient H-bonded chains. Since the dipole moment of individual OH groups sums up along such a transient chain, this process is dominant in dielectric spectroscopy, i.e. in the collective reorientation of permanent dipole moments. On the other hand, if local dielectric quantities are measured (like in solvation dynamics) or the measurement is sensitive to other molecular properties (like in depolarized dynamic light scattering), this Debye process plays a minor role. Thus, with the mentioned techniques, different dynamic aspects of the same sample can be investigated and only by combining them all a conclusive overall picture can emerge.

Characteristic in such mixtures is the large discrepancy between the dynamics of solvent and macromolecules, as typically the former move faster than the latter by several orders of magnitude.In extreme cases this even results in two distinct glass-transition temperatures, as resolved e.g. in calorimetric measurements. Data from dielectric spectroscopy usually display two distinct peaks, resulting from the dynamics of solvent and macromolecules, respectively (see image).

The influence of the macromolecules significantly alters the behavior of the solvent compared to the pure liquid. This results from the fact that from the perspective of the solvent, the macromolecules behave quasi rigid and therefore appear as a kind of matrix, confining the motion of solvent molecules. The matrix leads to enhanced dynamic heterogeneity, i.e. the distribution of solvent relaxation times is particularly broad, since the relaxation time of one specific solvent molecule significantly depends of its location with respect to the matrix. Additionally, interesting effects arise from macromolecule-solvent interactions, e.g. specific orientations of solvent molecules are favoured close to the matrix.

The limit of an infinitely slow matrix is referred to as hard confinement. It can be realized by filling a liquid into sponge- or tube-like pores. In such cases, liquids behave significantly different than in the bulk state. Their exact properties strongly depend on the diameter and the surface of the pores and sometimes a glassy state can be prepared in small pores, which is inaccessible otherwise,like in the case of nanoconfined water.

Their high ionic conductivity also makes them interesting for energy applications, for example as electrolytes. The sheer number of possible combinations of anions and cations – estimated to be about 10¹⁸ – makes a trial-and-error principle impractical, so a deeper understanding of ion dynamics is necessary to find the ideal ionic liquid for an application. We therefore investigate the interplay of ion transport and self-assembly, because often ionic liquids form domains consisting of polar and nonpolar molecular groups due to their chemical structure.

However, since liquid electrolytes should ideally be replaced by solid ones in the future, e.g. to eliminate the risk of leakage, we are experimenting with so-called ionogels. These consist of an ionic liquid and a solid matrix, such as gelatin or silica. The figure shows a gelatin-based ionogel, which has the consistency of a jelly, but is still an excellent ion conductor. This is demonstrated by the glowing LED, which is connected to a battery via the ionogel. The influence of the solid matrix on the dynamics of the ions is the main research question here to find the ideal balance of high conductivity and high mechanical stability of the ionogel.

Triplet state solvation dynamics (TSD) can be thought of as ”local dielectric spectroscopy with optical methods”. We are mainly concerned with extending the accessible time window, in which local information becomes available. While up until recently the time window was restricted to 1 ms – 1 s, we were able to extend the time range to 10⁻⁴ − 10 s by making use of new dye molecules. The left hand side of the figure shows the local relaxation of methyl-THF as an example.

In photon correlation spectroscopy (PCS), which is a time-domain light scattering technique, our development work mainly aims at improving the signal to noise ratio and getting rid of unwanted artifacts in the signal. In this way we are so far able to Fourier transform the measured correlation functions and combine these data sets with data sets from other light scattering techniques even for weakly scattering molecules. Thus, light scattering truly becomes a dynamic broadband technique.The right hand side of the figure shows a combined PCS and TFPI data set of glycerol that was obtained in that way.