Journal of Electroceramics 1:1, 65±71, 1997 # 1997 Kluwer Academic Publishers, Boston. Manufactured in The Netherlands. Characterization of Electrical Materials, Especially Ferroelectrics, by Impedance Spectroscopy ANTHONY R. WEST, DEREK C. SINCLAIR & NAOHIRO HIROSE University of Aberdeen, Department of Chemistry, Meston Walk, Aberdeen, UK AB24 3UE Received September 13, 1996; Accepted December 5, 1996 Abstract. A review is given of some of the problems encountered in the analysis and interpretation of impedance data. The importance of choosing the correct equivalent circuit to represent the data is emphasized and it is shown how ferroelectric materials, with their characteristic temperature-dependent capacitance, are particularly suited to discriminating between plausible equivalent circuits. Results are discussed for two materials, LiTaO 3 single crystal and BaTiO 3 ceramics. Keywords: impedance data, equivalent circuits, ferroelectric materials, impedance and modulus spectroscopy, lithium tantalate ferroelectric, barium titanate Introduction Ferroelectric materials, especially ceramics, are usually characterized by ®xed frequency measure- ments, typically at 1 kHz. From a practical point of view, this may be appropriate for measuring the key properties of loss tangent and dielectric constant or permittivity. From the viewpoint of scienti®c under- standing, however, such measurements instead of variable frequency measurements, place severe lim- itations on the amount of information that is readily obtainable, especially with materials that may be electrically heterogeneous. Electrical Microstructure and Impedance Spectroscopy The ready availability of impedance spectroscopy instrumentation covering, routinely, frequencies from millihertz to megahertz offers the possibility of a comprehensive characterization of electroceramics and in particular, characterization of their electrical microstructure. The concept of microstructure is, of course, fundamental to ceramics but the link between microstructure and electrical properties is not always obvious or straightforward. For instance, from a microscopic examination of ceramic texture, it is not usually possible to say whether the electrical proper- ties of grain boundaries are likely to be similar to or signi®cantly different from those of the individual grains. This may depend on whether the electrical properties are intrinsic to the material and are insensitive to minor variations in structure and composition or whether the properties are extrinsic since they could then vary dramatically if impurities are present. In the former case, grains and grain boundaries may be indistinguishable electrically, whereas in the latter their properties may differ greatly, especially if, as often happens, impurities segregate to the grain boundaries. A great strength of impedance spectroscopy is that, with appropriate data analysis, it is often possible to characterize the different electrically-active regions in a material, both qualitatively by demonstrating their existence and quantitatively, by measuring their individual electrical properties. Equivalent Circuits: Are They Necessary? Our belief is that identi®cation of the most appropriate equivalent circuit to represent the electrical properties of a material is absolutely essential as a step towards a