COLOSSAL DIELECTRIC CONSTANTS

1. Introduction

The development of new materials with high dielectric constants ε' is prerequisite for the further miniaturization of electronic devices. By broadband dielectric spectroscopy we aim to clarify the microscopic mechanisms that can lead to an enhancement of this material parameter.

The dielectric constant is a measure of the electrical polarizability of a material and the most important quantity in the construction of capacitive elements in electronics. Higher ε' allows for smaller dimensions of these elements. This is not only important for electronic-circuit elements, but can be also used for the enhancement of the energy and power density of very large capacitors that are employed instead of batteries for the short-term storage of energy, e.g., in hybrid cars. The capacitive materials in current electronics often have relatively low values of ε' of the order of 10.

The term "colossal dielectric constant" was first coined by our group [1] and denotes values of ε' > 1000. They are found in various materials and can be caused by different physical processes, both intrinsic and non-intrinsic.

For a review about colossal dielectric constants and the different physical mechanisms leading to them, see:

2. Mechanisms

a) Ferroelectricity

In ferroelectrics, very high ("colossal") values of ε' (10³-10⁵) arise from the ordering of dipolar degrees of freedom. Ferroelectric materials can be employed for the fabrication of very large capacitors. However, they have a number of disadvantages as, e.g., a strong temperature dependence of ε'. Aside of any application, the occurrence of ferroelectricity is a fascinating physical phenomenon, which has a long history of investigation in our group. Click here to learn more.

b) Maxwell-Wagner relaxations

Non-intrinsic effects can give rise to colossal dielectric constants via the long-known Maxwell-Wagner (MW) polarization, which arises from charge accumulation at interfaces. Interfaces of any kind can generate very high values of the dielectric constant because they act as parallel-plate capacitors with very small plate distances, thus having high capacitances. Examples are the depletion zones of Schottky diodes, forming at metallic contacts to semiconducting materials, grain boundaries in polycrystalline samples, so-called blocking electrodes arising at the surface of ionic conductors, and internal barriers in biological matter (e.g., cell walls). These effects lead to MW relaxations causing a strong frequency-dependence of the dielectric properties, which can be modelled by equivalent circuits [12]. MW effects are often misinterpreted to mirror intrinsic properties as dipole dynamics, ferroelectricity, etc., although they are completely non-intrinsic [2,13]. For a proper evaluation procedure of MW relaxations using an advanced equivalent-circuit approach, see:

We used a funny example, the dielectric measurement of a banana, to illustrate the occurrence of colossal dielectric constants and of P(E) hysteresis loops (like in a ferroelectric) due to purely non-intrinsic effects. We show, how the non-intrinsic origin of such effects can be easily identified by performing simple experiments:

Bananas go paraelectric A. Loidl, S. Krohns, J. Hemberger, and P. Lunkenheimer, J. Phys.: Condens. Matter 20, 191001 (2008).

c) Charge density waves

In charge-density-wave (CDW) materials, the charge density is a periodic function of position. Their dielectric behaviour shows two characteristic features: A harmonic-oscillator mode at GHz frequencies, caused by the CDW being pinned at defects, and a huge relaxation mode at kHz-MHz involving extremely large values of the dielectric constant, whose true origin is not completely clarified. As these materials have high dielectric losses, they cannot be applied in capacitive devices. An investigation of a CDW-related material showing colossal dielectric constants can be found in:

d) Hopping conductivity

Charge transport of localized charge carriers takes place via hopping conductivity, which is the most common conductivity mechanism in condensed matter. It is typical for any kind of disordered matter like amorphous or dirty semiconductors, glasses or ionic conductors. Hopping conductivity leads to a characteristic frequency dependence of the complex conductivity σ" + i σ", namely an approximate power-law increase: σ' ∝ σ" ∝ ν^s with s < 1. As ε' ∝ σ"/ν, this will always result in a divergence of ε'(ν) for low frequencies. However, the dielectric loss of such materials is relatively high (which of course is reasonable for a conducting material) rendering this effect unsuited for application in capacitor materials. Some relevant publications from our group in this field are:

3. Examples:

a) New colossal-dielectric-constant materials

In recent times, there have been strong efforts to find new high-ε' materials with better properties. The most prominent of them is CaCu₃Ti₄O₁₂ (CCTO). Compared to ferroelectrics-based dielectrics, CCTO exhibits a nearly temperature-independent colossal dielectric constant around room temperature, which is highly advantageous for possible technical application. While early works suggested intrinsic mechanisms to explain the high dielectric constant of this material, nowadays it is clear that it is of Maxwell-Wagner origin [2,6]. Aside of the famous CCTO, there are many other transition-metal oxides with colossal dielectric constant [1,8]. An example investigated by us is La_2-xSr_xNiO₄ which retains its colossal magnitude of ε' up to much higher frequencies [7,11].

b) Ionic conductors

In all ionic conductors, the dielectric constant strongly increases at low frequencies. This is due to the fact that the ions cannot penetrate into the metallic electrodes, leading to thin layers with immobile ions at the sample surface. These layers act as large capacitors and, thus, the dielectric constant increases at low frequencies [12]. This principle is used for the so-called supercapacitors which can store sizable amounts of energy and, in principle, have many advantages compared to batteries. However, to replace batteries by supercapacitors on a large scale, further development is necessary. In our group, fundamental research on new classes of ionic conductors is performed, which all show these "blocking electrodes". Click here to learn more.

4. Some relevant publications from our group: