Impedance spectroscopic study on well-defined (La,Sr)(Co,Fe)O3−δ model electrodes

https://doi.org/10.1016/j.ssi.2006.02.045Get rights and content

Abstract

Geometrically well-defined, dense thin film microelectrodes of the mixed conducting solid oxide fuel cell cathode material La0.6Sr0.4Co0.8Fe0.2O3−δ have been prepared by pulsed laser deposition and standard photolithographic techniques on yttria-stabilised zirconia substrates. The electrochemical properties of these model electrodes were investigated by impedance spectroscopy as a function of temperature and dc bias. It is shown that an equivalent circuit derived rigorously in Ref. [J. Jamnik and J. Maier, Phys. Chem. Chem. Phys. 3, 1668–1678 (2001).] for a cell with a mixed conducting electrode provides an appropriate description for this experimental system, enabling a correct interpretation of the measured impedance data. Under zero or small dc bias, the electrochemical resistance is dominated by the oxygen exchange reaction at the surface of the electrode, with minor contributions from the electrode/electrolyte interface and the ohmic resistance of the electrolyte. The main capacitive process is associated with oxygen stoichiometry changes in the bulk of the electrode (chemical capacitance), while an additional electrode/electrolyte interfacial capacitance is also present. The temperature and dc bias dependencies of these processes are discussed in terms of defect chemistry.

Introduction

Solid oxide fuel cells (SOFCs) combine high efficiency with flexibility in terms of the nature of the fuel. While low temperature polymer-electrolyte membrane (PEM) fuel cells are usually restricted to the conversion of purified hydrogen or methanol, the high temperature SOFC is much more tolerant in this respect and may as well be operated with natural gas, biogas, diesel or other fuels. The theoretical efficiency, defined as the ratio of Gibbs energy, ΔrG, and enthalpy, ΔrH, of the cell reaction, is high for typical SOFC reactions (e.g. 0.83 for H2 + ½O2  H2O(g) at 500 °C) and can in cases with a positive reaction entropy, ΔrS, even exceed one, which implies a cooling of the environment [2].

The efforts over the last decades, however, have shown that in spite of its conceptual attractiveness SOFC technology is also quite demanding. The actual performance and long term stability of existing SOFC systems are not satisfactory yet, and they are far from being economically competitive. One fundamental reason why the development of better and/or cheaper SOFCs is difficult lies in the complexity of the electrochemical processes involved. A prominent example is the oxygen reduction reaction at the cathode, which is often the main limiting factor to the performance of the whole SOFC system, especially at lower operating temperatures. Numerous studies on the electrochemical properties of different cathode materials can be found in the literature, but the mechanistic understanding of the oxygen reduction reaction is still unsatisfactory.

This lack of fundamental knowledge may be related to the fact that the vast majority of research done in this field has been performed on porous electrodes (e.g. [3], [4], [5], [6], [7], [8], [9], [10]). While porous systems have the advantage of being “realistic” from an application point of view, they suffer from the disadvantage of a usually ill-defined structure and geometry. This makes it very difficult to separate, for example, the influence of the electrode morphology from intrinsic properties of the material. In order to avoid these problems, several authors have investigated dense thin film electrodes [11], [12], [13], [14], [15], [16], [17].

Alternatively, dense thin film microelectrodes with a lateral dimension of 20–100 μm and typically 100 nm thickness, deposited on a single crystal of the electrolyte material, may be used. This experimental arrangement has a number of advantages: The microelectrodes are characterised by a well-defined and easily reproducible geometry and crystal orientation. They offer a large degree of experimental flexibility in so far as several hundred individual electrodes can be placed on each sample. This enables statistical averaging over a number of measurements in reasonable time, as well as a systematic investigation of irreversible effects which occur, for example, after the application of a large dc bias [18]. One further benefit of the microelectrode arrangement is that a reference electrode can be omitted in electrochemical experiments when working with an extended counter-electrode. Due to the large size-difference between working- and counter-electrode the electrochemical resistance of the latter is usually negligible. Finally, as will be shown later in this paper, the system is sufficiently simple such that it can unambiguously be modelled by an equivalent circuit, allowing for a reliable interpretation of experimental data.

The potential of dense thin film microelectrodes for basic studies of electrochemical electrode reactions has been demonstrated previously, e.g. in investigations on the oxygen reduction kinetics of La0.8Sr0.2MnOδ electrodes [19], [20], or by the discovery and characterisation of a strong electrochemical activation effect of La0.6Sr0.4Co0.8Fe0.2O3−δ electrodes [18].

In this work, it will be shown that detailed information on the oxygen reduction reaction can be obtained from temperature- and dc bias-dependent impedance measurements on thin film microelectrodes. La0.6Sr0.4Co0.8Fe0.2O3−δ (LSCF) has been chosen as typical mixed conducting cathode material with well-known high electrode performance [21], [22], [23], which is related to beneficial materials properties such as very high electronic and ionic conductivity [24], [25], [26], and fast oxygen surface exchange [25], [27].

Section snippets

Sample preparation and characterisation

Dense films of La0.6Sr0.4Co0.8Fe0.2O3−δ were fabricated in a first step by pulsed laser deposition (PLD) on polished, 9.5 mol% Y2O3-doped ZrO2 single crystals (CrysTec, Germany) with (100)-orientation and dimensions 5 × 5 × 0.5 mm. A commercially obtained La0.6Sr0.4Co0.8Fe0.2O3−δ target (HITEC Materials, Germany) was irradiated by 248 nm UV light from a KrF excimer laser with an energy density per pulse of 1.6 J/cm2 at a repetition frequency of 5 Hz. During ablation the substrate temperature was

Model used for the interpretation of impedance spectra

Equivalent circuit representations consisting of resistors, capacitors and other elements are often employed to evaluate and interpret experimental impedance data. Frequently, these models are proposed ‘ad hoc’ and tested in terms of their apparent agreement with the measured spectra. Intuitively, a good correspondence between experimental data and fitted curve may then be taken as a confirmation that the proposed circuit provides a good description of the system under investigation. Such a

Sample characterisation

In the optical microscope, the 100 nm thin LSCF microelectrodes are clearly visible as dark circles (Fig. 2a). The surface topography of as-prepared electrodes has been characterised by white light interferometry (Fig. 2b) and atomic force microscopy (Fig. 3a). The lateral extension of the “hills” in the topographic image of Fig. 3a indicates a grain size of the order of 100 nm. The average surface roughnessRa=1Ni=1N|ZiZ¯|(Zi: height of a point i on the surface, : average height) obtained

Conclusion

It has been demonstrated that dense thin film microelectrodes of the mixed conducting material La0.6Sr0.4Co0.8Fe0.2O3−δ on a YSZ electrolyte can be employed as a well-defined model system to obtain mechanistic information on the oxygen reduction reaction at SOFC cathodes. An adapted version of an equivalent circuit derived in Ref. [1] has been shown to be an appropriate model for this experimental system. The interpretation of the experimentally observed impedance data according to the

Acknowledgements

The authors thank G. Cristiani and B. Stuhlhofer for preparing the samples used for this study. Financial support of Deutsche Forschungsgemeinschaft (DFG priority program 1060) and German–Israeli Foundation (GIF-project G-703.41.10/2001) is gratefully acknowledged.

References (45)

  • J.-M. Bae et al.

    Solid State Ionics

    (1998)
  • B.C.H. Steele et al.

    Solid State Ionics

    (1998)
  • V. Dusastre et al.

    Solid State Ionics

    (1999)
  • S.P. Jiang

    Solid State Ionics

    (2002)
  • A. Endo et al.

    Solid State Ionics

    (2000)
  • A. Ringuedé et al.

    Solid State Ionics

    (2001)
  • M. Sase et al.

    J. Phys. Chem. Solids

    (2005)
  • V. Brichzin et al.

    Solid State Ionics

    (2002)
  • C. Ftikos et al.

    J. Eur. Ceram. Soc.

    (1993)
  • S. Wang et al.

    Solid State Ionics

    (2003)
  • M. Katsuki et al.

    Solid State Ionics

    (2003)
  • R.A. De Souza et al.

    Solid State Ionics

    (1998)
  • J. Fleig

    Solid State Ionics

    (2002)
  • B.C.H. Steele

    Solid State Ionics

    (1995)
  • H.J.M. Bouwmeester et al.

    Solid State Ionics

    (1994)
  • C.H. Chen et al.

    Solid State Ionics

    (1997)
  • J. Jamnik et al.

    Electrochim. Acta

    (1999)
  • J. Fleig et al.

    J. Eur. Ceram. Soc.

    (2004)
  • J. Maier

    Solid State Ionics

    (1998)
  • W. Sitte et al.

    Solid State Ionics

    (2002)
  • L.-W. Tai et al.

    Solid State Ionics

    (1995)
  • M. Filal et al.

    Solid State Ionics

    (1995)
  • Cited by (447)

    View all citing articles on Scopus
    View full text