Mobile charges and lattice polarization interact in ferroelectric materials because of the Coulomb interaction between the mobile free charges and the fixed lattice dipoles. We have investigated this mutual screening in KTiOPO4, a ferroelectric/super
ionic single crystal in which the mobile charges are K+ ions. The ionic accumulation close to the crystal surfaces leads to orders of magnitude increase of the Second Harmonic Generation. This ionic space charge model is supported by the absence of such an effect in non-ionic conductor but ferroelectric BaTiO3, by its temperature dependence in KTiOPO4 and by its broad depletion at domain walls.
Artificial tuning of dielectric parameters can result from interface conductivity in polycrystalline materials. In ferroelectric single crystals, it was already shown that ferroelectric domain walls can be the source of such artificial coupling. We s
how here that low temperature dielectric losses can be tuned by a dc magnetic field. Since such losses were previously ascribed to polaron relaxation we suggest this results from the interaction of hopping polarons with the magnetic field. The fact that this losses alteration has no counterpart on the real part of the dielectric permittivity confirms that no interface is to be involved in this purely dynamical effect. The contribution of mobile charges hopping among Fe related centers was confirmed by ESR spectroscopy showing maximum intensity at ca Tsim40 K.
This review will deal with several types of free charge localisation in oxides and their consequences on the effective dielectric spectra of such materials. The first one is the polaronic localisation at the unit cell scale on residual impurities in
ferroelectric networks. The second one is the collective localisation of free charge at macroscopic interfaces like surfaces, electrodes and grain boundaries in ceramics. Polarons have been observed in many oxide perovskites mostly when cations having several stable electronic configurations are present. In manganites, the density of such polarons is so high as to drive a net lattice of interacting polarons. On the other hand, in ferroelectric materials like BaTiO3 and LiNbO3, the density of polarons is usually very small but they can influence strongly the macroscopic conductivity. The contribution of such polarons to the dielectric spectra of ferroelectric materials is described. Even residual impurities as for example Iron can induce well defined anomalies at very low temperatures. This is mostly resulting from the interaction between localised polarons and the highly polarisable ferroelectric network in which they are embedded. The case of such residual polarons in SrTiO3 will be described in more details, emphasizing the quantum polaron state at liquid helium temperatures. Recently, several non-ferroelectric oxides have been shown to display giant effective dielectric permittivity. It is first shown that the frequency/temperature behaviour of such parameters is very similar in very different compounds (donor doped BaTiO3, CaCu3Ti4O12, LuFe2O4,Li doped NiO,...). This similarity calls for a common origin of the giant dielectric permittivity in these compounds. A space charge localisation at macroscopic interfaces can be the key for such extremely high dielectric permittivity.
The recent quest for improved functional materials like high permittivity dielectrics and/or multiferroics has triggered an intense wave of research. Many materials have been checked for their dielectric permittivity or their polarization state. In t
his report, we call for caution when samples are simultaneously displaying insulating behavior and defect-related conductivity. Many oxides containing mixed valent cations or oxygen vacancies fall in this category. In such cases, most of standard experiments may result in effective high dielectric permittivity which cannot be related to ferroelectric polarization. Here we list few examples of possible discrepancies between measured parameters and their expected microscopic origin.
Triggered by the revival of multiferroic materials, a lot of effort is presently undergoing as to find a coupling between a capacitance and a magnetic field. We show in this report that interfaces are the right way of increasing such a coupling provi
ded free charges are localized on these two-dimensional defects. Starting from commercial diodes at room temperature and going to grain boundaries in giant permittivity materials and to ferroelectric domain walls, a clear magnetocapacitance is reported which is all the time more than a few percent for a magnetic field of 90kOe. The only tuning parameter for such strong coupling to arise is the dielectric relaxation time which is reached on tuning the operating frequency and the temperature in many different materials.
