Parametric sensitivity analysis on the cold start process of a polymer electrolyte membrane fuel cell

Subzero start-up of the polymer electrolyte membrane fuel cell (PEMFC) is one of the most challenging tasks to be solved before commercialization. During the subzero start-up process, water generated in the oxygen reduction reaction at the cathode side of PEMFC is susceptible to freeze?which makes active sites covered by the ice and gases failed to reach the surface of the catalyst layer (CL), leading to a substantial decay and even ending up with a failure of the start-up. Given that many factors affect the cold start process, the relative contribution of the essential factors on the cold start process is independently analyzed using first-order finite-difference sensitivity analysis from -20 degrees C to -30 degrees C. The effect of essential parameters on the cold start process is quantified. The investigated parameters include the ratio of bipolar plate (BP) thickness to that of the CL (R-BP/CL), the starting voltage (V-ini), stoichiometry ratios, inlet gas temperature (T-in), the porosity of the CL (epsilon(CL)), initial membrane water content (lambda(ini)), membrane thickness (theta(mem)), the volume fraction of ionomer (omega(CL)) in CLs, and the heat capacity of the BP (rho c(pBP)). Results show that the cold start process is most sensitive to R-BP/CL and lambda(ini). Significant improvement of cold start performance can be achieved by appropriately adjusting R-BP/CL, lambda(ini), T-in, and epsilon(CL). Appropriately increasing the V-ini also can be a method to improve cold start performance, especially for the cold start from -30 degrees C. Besides, optimized omega(CL) in CLs, proper theta(mem), and lower rho c(pBP) can contribute to better performance, especially for the cold start from -20 degrees C.

Automated summary

The subzero start-up of PEMFC faces challenges due to water freezing during the oxygen reduction reaction, affecting the active sites and gas diffusion. Sensitivity analysis showed that the cold start process is most influenced by R-BP/CL and lambda(ini), and can be improved by adjusting these factors. Additionally, adjusting V-ini, T-in, epsilon(CL), omega(CL), theta(mem), and rho c(pBP) can also contribute to better cold start performance.