• A systematic study of HT-PEMFC response to realistic methanol reformate compositions • Reformate gas compositions based on methanol steam reforming 0D process modeling. • Spatial mapping reveals impurity effects and how water vapor restores current distribution. • DRT analysis separates impurity effects on electrochemical processes. HT-PEMFCs are attractive for operation with methanol steam reforming systems, but their response to individual reformate components remains complex and spatially non-uniform. This study investigates the effects of carbon dioxide (CO₂), carbon monoxide (CO), and water vapor (H₂O) on HT-PEMFC performance under representative reformate compositions at 160°C and 170°C. Gas mixtures containing hydrogen with controlled CO₂ (10-30%) and CO (0.1-3%) concentrations were examined, while water vapor levels (1.0-2.5% RH) were selected based on reformer process modeling. A combined diagnostic approach using polarization curves, electrochemical impedance spectroscopy with distribution of relaxation times (DRT), and segmented current density mapping was applied. CO₂ induces performance losses primarily through hydrogen dilution and changes in fast anode processes, with higher overall degradation observed at 170°C due to increased ohmic resistance rather than kinetic. CO addition causes severe voltage losses, particularly at 160°C, driven by catalyst site blocking and strong kinetic limitations. Spatial analysis reveals that both CO₂ and CO preferentially suppress initially highly active MEA regions, leading to a more uniform-but overall reduced-current distribution. Introducing water vapor improves performance mainly by reducing the ohmic resistance associated with membrane hydration. This effect is spatially localized near the anode inlet and does not fully restore previously poisoned regions, indicating that hydration mitigates transport and conductivity losses but does not reverse kinetic deactivation. These findings highlight the importance of temperature-dependent ohmic effects and spatial diagnostics for realistic reformate-fed HT-PEMFC operation.
Todorovski et al. (Sun,) studied this question.