Localised Electrochemical Impedance Measurement Using an Innovative Reference Electrode Array

Thursday, 30 July 2015: 08:40
Dochart (Scottish Exhibition and Conference Centre)
E. Brightman (National Physical Laboratory), E. Engebretsen (University College London), C. Lyu (Harbin Institute of Technology), and G. Hinds (National Physical Laboratory)
Electrochemical impedance spectroscopy (EIS) is an increasingly popular diagnostic tool for polymer electrolyte membrane fuel cells (PEMFCs) and other electrochemical energy conversion devices. Data are commonly fitted to an equivalent circuit model enabling empirical determination of overpotentials, double layer capacitance, and other key parameters. However, conventional experimental configurations are limited to two-electrode measurements, which rely on the assumption that the anode contribution to impedance is negligible. An additional limitation is that an average response for the whole cell is obtained, whereas it is generally accepted that local inhomogeneity in reactant concentration, water distribution and catalyst utilisation is a significant factor affecting fuel cell durability. The use of segmented cells to obtain a functional map of fuel cell performance is well established, but their application to impedance spectroscopy requires complicated equipment and expensive hardware.

Here we report the first localised EIS measurements in a single cell PEMFC using an innovative reference electrode array developed at NPL. By making contact directly with the electrode of interest using a Nafion tube salt bridge inserted through the end plates of the cell, the NPL reference electrode gives an iR-free, localised potential measurement, with the advantage that commercial MEAs can be tested directly with minimal modification to cell hardware. An array of reference electrodes is used to measure the spatial variation in impedance across the active area of the cell under a range of operating conditions, before and after accelerated aging of a commercial MEA. The advantages and disadvantages of this technique and the implications for modelling and optimisation of fuel cell design are discussed.