Bernhard Roling

Fundamental questions relating to ion conduction in disordered solids

A number of basic scientific questions relating to ion conduction in homogeneously disordered solids are discussed. The questions deal with how to define the mobile ion density, what can be learnt from electrode effects, what the ion transport mechanism is, the role of dimensionality and what the origins of the mixed-alkali effect, the time-temperature superposition, and the nearly constant loss are. Answers are suggested to some of these questions, but the main purpose of the paper is to draw attention to the fact that this field of research still presents several fundamental challenges.

Li10SnP2S12: an affordable lithium superionic conductor.

The reaction of Li2S and P2S5 with Li4[SnS4], a recently discovered, good Li(+) ion conductor, yields Li10SnP2S12, the thiostannate analogue of the record holder Li10GeP2S12 and the second compound of this class of superionic conductors with very high values of 7 mS/cm for the grain conductivity and 4 mS/cm for the total conductivity at 27 °C. The replacement of Ge by Sn should reduce the raw material cost by a factor of ~3.

Scaling properties of the conductivity spectra of glasses and supercooled melts

The ion transport properties of glasses and supercooled melts are studied using the scaling properties of complex conductivity data. It is shown that the frequency response of the conductivity reflecting the transport mechanism is very similar in glasses and in melts. In the crossover regime from dc to dispersive conductivity, the frequency response depends on composition, while a universal response is found at higher frequencies. The outcomes of an alternative analysis based on the complex electrical modulus formalism are examined in detail, and it is demonstrated that the shape of the M″ peak is not significant with respect to the ion transport mechanism.

Enhanced lithium transference numbers in ionic liquid electrolytes.

Ion transport processes in mixtures of N-butyl- N-methyl-pyrrolidinium bis(trifluoromethanesulfonyl)imide (BMP-TFSI) and lithium bis(trifluoromethanesulfonyl)imide (Li-TFSI) were characterized by ac impedance spectroscopy and pulsed field gradient NMR. Molar ratios x = n Li-TFSI/( n Li-TFSI + n BMP-TFSI) up to 0.377 could be achieved without crystallization. From the bulk ionic conductivity and the individual diffusion coefficients of cations and anions we calculate the Haven ratio and the apparent lithium transference number. Although the Haven ratio exhibits typical values for ionic liquid electrolytes, the maximal apparent lithium transference number is higher than found in other recent studies on ionic liquid electrolytes containing lithium ions. On the basis of these results we discuss strategies for further improving the lithium transference number of such electrolytes.

What do electrical conductivity and electrical modulus spectra tell us about the mechanisms of ion transport processes in melts, glasses, and crystals?

Abstract The significance of electrical conductivity and electrical modulus spectra with respect to ion transport mechanisms in melts, glasses, and crystals is discussed. It is shown that in the typical frequency range of an ac impedance experiment, the real part of the conductivity, σ′, is exclusively determined by the ion transport dynamics, while the imaginary part of the modulus, M″, is additionally influenced by vibrational and electronic polarisations. Accordingly, conclusions from the shape of M″ spectra about ion transport mechanisms cannot be drawn without properly taking into account the influence of these polarisations. Generally, it can be argued that analyses in terms of conductivity or dielectric spectra are preferable to analyses in terms of modulus or resistivity spectra if the electrical properties of a sample can be represented by resistances acting in parallel.

Dynamics of mobile ions in crystals, glasses and melts

Abstract In crystalline, glassy and molten electrolytes, the dynamics of the mobile ions is best studied by taking conductivity spectra, i.e., by measuring ionic conductivities in the entire range from dc to far-infrared frequencies. Experimental conductivity spectra are reviewed and discussed, with particular emphasis on their high-frequency plateaux and their low-frequency scaling properties. The concept of mismatch and relaxation (CMR) is shown to provide a general basis for understanding the spectra in terms of the ion dynamics. The conductivity spectra of crystalline fast ion conductors are, e.g., explained by the jump relaxation model, which builds on the CMR. In glassy electrolytes, the mobile ions encounter different kinds of site, and the jump relaxation model has to be modified accordingly. Formulating the CMR without reference to the existence of fixed sites yields a description of the ion dynamics in simple molten salts. The model is thus able to reproduce both the conductivity spectra of fragile molten salts and the characteristic temperature dependence of their dc values. The remarkable low-frequency scaling behaviour of glassy electrolytes is considered a manifestation of the particular long-time properties of the CMR.

Mixed alkaline-earth effects in ion conducting glasses

Abstract We report ionic mobilities in silicate glasses containing sodium oxide and either one or two alkaline–earth oxides, based on dynamic mechanical thermal analysis (DMTA) and electrical conductivity spectroscopy. In mixed alkaline–earth glasses, the mobilities of alkaline–earth ions are considerably lower than in the corresponding single alkaline–earth glasses. Furthermore, the glass transition temperatures of mixed alkaline–earth glasses are lower than expected from a linear extrapolation of the transition temperatures of the corresponding single alkaline–earth glasses. In order to quantify the degree of decoupling of the mobile ions from the glassy network, we define a new decoupling ratio which makes use of both mechanical and electrical data. We, thereby, show that the effect of mixing divalent ions is qualitatively similar to the effect of mixing monovalent ions. Quantitatively, the mixed alkaline–earth effect is less pronounced than the mixed alkali effect. A possible explanation for this observation is given.

Hysteresis Effects in the Potential-Dependent Double Layer Capacitance of Room Temperature Ionic Liquids at a Polycrystalline Platinum Interface

This paper has been withdrawn by the author since it had been submitted to Journal of Physical Chemistry C.

Information on the absolute length scales of ion transport processes in glasses from electrical conductivity and tracer diffusion data

We show that absolute values for the time-dependent mean-square displacement of mobile ions in glasses can be obtained from a combined analysis of electrical conductivity and tracer diffusion data. By applying this method to sodium germanate glasses of different compositions, we find that the distances the ions cover in characteristic stages of the transport process increase considerably with decreasing sodium oxide content. At low sodium oxide contents, we observe a linear relation between the transport distances and the interionic separation distances, while this relation is lost at high sodium oxide contents. The same is found to be true for other alkali ion conducting glassy systems. The analysis of this effect is important for discriminating between the influence of the network structure and of the interionic Coulomb interactions on the ion dynamics.

Slow and fast capacitive process taking place at the ionic liquid/electrode interface

Electrochemical impedance spectroscopy was used to characterise the interface between the ultrapure room temperature ionic liquid 1-butyl-1-methylpyrrolidinium tris(pentafluoroethyl)trifluorophosphate and a Au(111) working electrode at electrode potentials more positive than the open circuit potential (-0.14 V vs. Pt pseudo-reference). Plots of the potential-dependent data in the complex capacitance plane reveal the existence of a fast and a slow capacitive process. In order to derive the contribution of both processes to the overall capacitance, the complex capacitance data were fitted using an empirical Cole-Cole equation. The differential capacitance of the fast process is almost constant between -0.14 V and +0.2 V (vs. Pt pseudo-reference) and decreases at more positive potentials, while the differential capacitance of the slower process exhibits a maximum at +0.2 V. This maximum leads to a maximum in the overall differential capacitance. We attribute the slow process to charge redistributions in the innermost ion layer, which require an activation energy in excess of that for ion transport in the room temperature ionic liquid. The differential capacitance maximum of the slow process at +0.2 V is most likely caused by reorientations of the 1-butyl-1l-methylpyrrolidinium cations in the innermost layer with the positively charged ring moving away from the Au(111) surface and leaving behind voids which are then occupied by anions. In a recent Monte Carlo simulation by Federov, Georgi and Kornyshev (Electrochem. Commun. 2010, 12, 296), such a process was identified as the origin of a differential capacitance maximum in the anodic regime. Our results suggest that the time scales of capacitive processes at the ionic liquid/metal interface are an important piece of information and should be considered in more detail in future experimental and theoretical studies.