Current state-of-the-art devices for energy conversion and storage still lack the efficiency required for overcoming the energy transition, which is one of society’s major challenges. Polycrystalline thin films with high electronic and/or ionic conductivity play a key role in many energy devices. To tailor the electrical properties of the thin films to the needs of their application,  careful material selection and optimized microstructure are necessary. Charge transport properties in new promising mixed conducting materials are typically characterized by impedance spectroscopy (IS). A small sinusoidal excitation signal is applied in the form of a current or a voltage, and the response, which is also sinusoidal, is measured. The frequency range will cover several orders of magnitude making it possible to distinguish between individual transport processes in the spectrum if their characteristic frequencies differ. Additionally the required basic measurement setup is relatively simple. This combination makes IS one of the most powerful and frequently used experimental tools in the field of solid-state ionics.Analyzing impedance spectra is usually accomplished by fitting the experimental data with the impedance of an equivalent circuit consisting of a few components only. This yields macroscopic transport quantities, such as resistances, capacitances and inductances, which are then qualitatively correlated to microscopic transport processes in the sample using theoretical models. The widely used model, the brick layer model, approximates polycrystalline thin films by a network of identical cubes based on the average grain size with a linear potential gradient. However, this model does not consider the impact of different transport paths, among other factors. The true effect of the microstructure on the impedance is still not well understood.

Our study demonstrates the significant influence of microstructure on the impedance of the sample. We  perform IS on polycrystalline ceria microstructures using microcontacts as electrodes. Measurements are taken at varying oxygen partial pressures and temperatures. To minimize any artifacts, a specially-built IS measurement setup is used. The microstructures, which are characterized using optical and scanning electron microscopy, have an extension of 20 x 50 µm² with grain sizes of several µm and provide a defined transport path network. Micro Raman spectroscopy is utilized to yield a spatially resolved material characterization. Computer-aided simulations based on an impedance network model are used to analyze the experimental IS data. The simulations accurately map and implement the ceria microstructure. The results indicate that the microstructure has a stronger influence on the impedance than previously thought. Existing models often oversimplify this influence, leading to incorrect assumptions of the microscopic properties. Further investigations are promising for establishing a new correlation model.