ElsevierGiner Sanz, JJ.; Ortega Navarro, EM.; PĂ©rez-Herranz, V. (2015). Optimization of the electrochemical impedance spectroscopy measurement parameters for PEM fuel cell spectrum determination. Electrochimica Acta. 174:1290Acta. 174: -1298Acta. 174: . doi:10.1016Acta. 174: /j.electacta.2015 Optimization of the electrochemical impedance spectroscopy measurement parameters for PEM fuel cell spectrum determination Keywords: Electrochemical Impedance Spectroscopy; measurement parameters; optimization; factorial experimental design; PEMFC.
AbstractCurrently, electrochemical Impedance Spectroscopy (EIS) is a widely used tool for the study of electrochemical systems, in general; and fuel cells, in particular. A great effort is typically invested in the analysis of the obtained spectra; whereas, little time is usually spent optimizing the measurement parameters used to obtain these spectra. In general, the default settings provided by the control software used to perform the measurements, or the parameters used in similar systems available in literature, are selected to carry out the measurements. The goal of this work is to determine the optimal measurement parameters for obtaining impedance spectra of a commercial PEM fuel cell. In order to achieve this, a 2 5 factorial design was considered. Five factors were considered, the five impedance spectroscopy measurement parameters: maximum integration time; minimum number of integration cycles; number of stabilization cycles; maximum stabilization time; and minimum cycle fraction. For each factor combination envisaged in the experimental design, the cell spectrum was obtained in given operation conditions, for which the reference spectrum of the system was known, since it had been determined in previous works. The experimentally obtained spectra were fitted to the reference electric equivalent circuit. The mean square error between the experimental data fitting and the reference spectrum fitting was determined in each case, and was used as the dependant variable for the experimental design analysis. An analysis of the variance was performed in order to determine which measurement parameters have a significant effect on the dependant variable; and a model relating the dependant variable and the measurement parameters was built. This model was used in order to obtain the optimal value of each one of the measurement parameters that minimized the mean square error of the fit obtained from the experimental data with respect to the reference fit.