The operating conditions in a PEMFC are by nature heterogeneous because the concentration in oxygen and hydrogen diminishes along the channels while gases enrich in water. Furthermore, depending on the quality of the cooling circuit, the temperature may not be uniform over the MEA surface. Although modeling can put forward some local variations in current density, gas concentration, and liquid water content in PEMFC [1-2], the diversity and complexity of the physical phenomena taking place in MEA makes experimental investigations indispensable. Numerous visualization techniques have been used, from optical photography [3, 4] and Magnetic Resonance Imaging [5] to neutron [6] and synchrotron X-ray imaging [7]. However, visualization techniques mostly provide qualitative information, so that quantitative information about local operation inside an elementary cell is generally obtained using segmented cells. Many groups have developed their own tools [8] but fundamentally, a segmented cell is an ordinary single cell with one of the electrodes divided into smaller electrodes, where current density, impedance or even potential and active surface area can be measured independently of the other segments. Segmented cells are increasingly used for in situinvestigation of PEM fuel cell operation, to design flow fields, or to optimize the performances [9]. In some cases, segmented cells can be used in parallel with a visualization technique, for instance to study the impact of channel flooding on the local current density [4]. Using segmented cells recently enabled understanding the origin of local degradations in PEMFC, in occasions put forward by post-mortem analyses [10]. One of the most spectacular phenomena they contributed to unveil is the reverse currents occurring during cell start-up or shut-down under open circuit conditions. Their main mechanisms were initially presented by Reiser et al. [11] but to the best of our knowledge, Siroma et al. [12] and Maranzana et al. [13] were the first to succeed in measuring internal currents in PEMFC. From our point of view, there are currently two main perspectives in the use of segmented cells. The first one is to explore degradation mechanisms in other situations than start-up and shut-down events, for instance when a fuel cell is operated in dead-end mode [14]: in this case, excessive accumulation of liquid water and possibly nitrogen and oxygen in the anode compartment can result in significant rise of the anode potentials. More generally, segmented cells could be used to analyze the local operation of a cell during any kind of transient or steady-state operation. The main challenge in these cases consists in finding reliable markers of the degradation of the electrolyte membrane: up to now, instrumented cells were mostly used to monitor the degradation of the electrodes. The second perspective concerns the temperature heterogeneity over the MEA surface and through its thickness. Contrary to local variations in gas concentration, the impact of temperature heterogeneities is far from being well understood although MEA components degradation phenomena, reaction mechanisms/kinetics, and mass-transport phenomena are all dependent on that parameter. These issues and the main dificulties in designing and operating segmented cells will be discussed in the presentation.
Read full abstract