Effect of composition on the performance of cermet electrodes. Experimental and theoretical approach

Abstract

This work aims to analyse the behaviour of cermet electrodes as a function of their composition, i.e. the ratio between ionic and electronic conducting particles. This is an important parameter to be considered to obtain maximum performance from this type of electrode, which is currently under study for application in oxygen sensors and solid oxide fuel cells. Experimental results of overall electrode resistance, including both ohmic and activation polarisation effects, have been obtained through electrochemical impedance spectroscopy measurements of Pt/YSZ electrodes in air. The results compare favourably with the theoretical predictions for several compositions above the percolation threshold of the electronic conductor. For this reason, the model is a useful tool for the design of optimised cermet electrodes; in particular, the experimental data show that maximum performance is attained for compositions very close to the percolation threshold of the electronic conductor, and this is in very good agreement with the modelling results.

Introduction

Cermet composite electrodes allow performance improvement with those electrochemical devices which employ a solid electrolyte, and this is the reason why they are currently under study for application in oxygen sensors and solid oxide fuel cells. Indeed, a low polarisation resistance is required for both applications: to gain satisfactory sensitivity even at low temperatures in the case of sensors [1], and to reduce the losses encountered in the energy conversion process in the case of solid oxide fuel cells [2], [3]. The reason for the superiority of cermet composite electrodes to traditional single-phase electrodes (formed of a porous layer of electron-conducting material) is that when the latter are coupled to a solid electrolyte the active area of the electrochemical reaction (which is related to the contact area between the electronic conductor and the ionic conductor) is confined to the interface between the electrode and the electrolyte. On the contrary, composite electrodes are formed by a mixture of electronic and ionic conducting materials, and thus the active area for the electrochemical reaction is extended three-dimensionally into the electrode thickness. In this way, composite electrodes show a polarisation resistance which can be even several order of magnitudes lower than that of traditional electrodes [4]. Since the electrochemical reaction occurs throughout the composite electrode, in a manner which is analogous to a porous electrode penetrated by a liquid electrolyte, the behaviour of the mixed electrode can be analysed on the basis of the single pore model [5]. The results can be summarised by the following formula [3], [6], for the overall electrode resistance R, accounting for both the ohmic and activation polarisation effects:

$$\text{R=}\text{ρχ}\text{coth}\text{ρa}{}^2\text{χ}$$

where a is the thickness of the electrode, χ is the local polarisation resistance and ρ is a parameter proportional to the resistivity of the ion-conductive phase in the composite electrode, the proportionality constant depending on the electrode morphology. This formula proved to be in good agreement with the experimental results of the measured polarisation resistance of electrodes of different thicknesses [3], [6]. However, other important parameters, such as the composition and the particle dimensions, were investigated. The effect of those two parameters has then been analysed through some purely theoretical studies, based either on analytical models, [7], [8] or on Monte-Carlo tools [8]. Both analytical and Monte-Carlo models demonstrate that only if the composition and morphology of the electrode are accurately chosen can the maximum electrode performance be obtained, and, in particular, both models emphasise the importance of the percolation thresholds. In fact, one of the most important requirements for correct operation of a composite electrode is that both the electron and the ion conduction paths are continuous throughout the electrode. Only in this way can the electrical charges involved in the electrochemical reaction effectively supply the whole electrode, and thus only in this way is the three-dimensional extension of the reaction area effective throughout the electrode. The continuity of the conduction paths in a mixed conductor is a percolation problem, and it is attained only if the volume fraction of the particles of each type is above a critical level, i.e. the percolation threshold. The electrode has two percolation thresholds, one for the electronic conductor, and another one for the ionic conductor; a sharp increase in the performance has been predicted to occur at the percolation thresholds and there is a window of compositions between the percolation thresholds in which the performance is predicted to be very high [7]. A fairly good agreement between the percolation thresholds of the Monte-Carlo and the analytical model has been demonstrated [8].

The literature reports several experimental studies on composite electrodes [9], [10], [11]. For example, Dees et al. [4] have investigated the electrical conductivity of Ni/ZrO₂–Y₂O₃ cermets at 1273 K, and their results show an improvement of several orders of magnitude with a volumetric fraction of Ni of about 30%, in agreement with the results of the percolation theory. Dusastre et al. [12] have studied the performance of composite La_1−xSr_xFe_1−yCo_yO_3−δ/Ce_0.9Gd_0.1O_2−δ (LSCF/CGO) cathodes. In this study, the importance of the percolation thresholds is discussed experimentally and theoretically; however, since LSCF is a mixed electronic and ionic conductor, the oxygen ion transport in the composite electrode and thus the electrode performance as a function of the electrode composition has different features from a cermet composite formed of a mixture of a pure electron and a pure ion conductors. Hu et al. [13] have analysed Ag/BaCe_9.8Gd_0.2O₃ electrodes with different compositions, and some of their experimental data are in roughly good agreement with the results of the percolation theory.

However, no experimental studies have been devoted to the analysis of the overall polarisation resistance of a cermet electrode as a function of its composition, and, indeed, a comparison between theoretical and experimental results, which is the aim of the present paper, has not been presented so far. Since the aim of this work is to demonstrate a correlation between microstructure and performance, rather than to find new electrode materials with high electrochemical activity, Pt/YSZ cathodes have been investigated, even if the electrochemical activity of Pt on oxygen reduction at high temperatures is known to be lower than that of other catalysts.