Effect of Cationic Contaminants on PEM Fuel Cell Performance

Resilience against cationic impurities is one of the major requirements for polymer electrolyte membrane fuel cells (PEMFC), especially in automotive applications. In order to investigate the role of cationic impurities and to correlate the uptake of cationic contaminants on the membrane electrode assembly (MEA) with fuel cell performance, an MEA (W.L. Gore, PRIMEA Series 57) (MEA#2) with 50 cm active area was contaminated by soaking in a solution of 28.26 mM H2SO4 and 0.58 mM Al2(SO4)3 in DI water following Kienitz’s procedure [1]. The MEA was then tested in a single fuel cell to investigate fuel cell performance as a function of cationic contamination level. An MEA coupon was also contaminated at the same time and was characterized to determine the cationic occupation fraction by acid site titration. As shown in Figure 1, available acid capacity and total acid capacity of the contaminated MEA (MEA #2) are similar to that for the non-contaminated MEA (MEA #1). It is calculated that the Al occupies ~8% of the active sites in the MEA by comparing the available acid capacity with total acid capacity for contaminated MEA. Figure 2 shows the performance of the MEA#2 during a current hold test and it is seen that the performance is very low. The cell voltage is ~0.4 V at a current density of 0.1 A/cm. For reference, under same operating conditions, these MEAs will show a cell performance of ~0.6 V at 1 A/cm. It is also seen that the resistance of the cell as measured with current interrupt (CI) is several orders higher than uncontaminated cells. In order to recovery the cell performance, the test fuel cell was disassembled, and the MEA is soaked in a 1 M H2SO4 solution for 24 hours at room temperature, and then washed with DI water. The procedure is similar to the total acid capacity titration procedure with the exception of the type of the acid, i.e. H2SO4 is used instead of HCl for better chemical compatibility with the MEA. The MEA is re-tested under identical conditions. As shown in Figure 3, the performance of the cell improved significantly. Although not shown here, the performance obtained after reprotonation is very similar to an as-received, uncontaminated MEA. Although the titration measurements indicate that the cationic occupation of the MEA is low, the performance loss is very significant. Further protonation of the MEA and ensuing recovery show that the MEA is otherwise healthy so the entire performance loss is due to cationic contamination. Kienitz [1] reported similar results: the decrease in performance was significant even with only 13% of the active sites contaminated by cesium cations. He contemplated that the fuel cell reaction might be hindered because there were fewer protons in the cathode available to participate in oxygen reduction. In addition, a gradient of protons across the cell might alter the thermodynamics of the system. We hypothesize that due to large potential gradient across the membrane during operation, the positively charged ions accumulate in the cathode catalyst layer and cathode side of the membrane, deactivating the active sites in cathode catalyst layer and causing the majority of the performance loss. Further experiments and computational modeling of the cation transport and contamination are underway to verify this theory.