skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: EFFECTS OF FUEL IMPURITIES ON PEM FUEL CELL PERFORMANCE.

Conference ·
OSTI ID:975330

Power generation with polymer electrolyte membrane fuel cells (PEMFC), particularly those designed for domestic and transportation applications, will likely operate on hydrogen reformed from hydrocarbons. The primary sources of H{sub 2} can be methane (from natural gas), gasoline or diesel fuel. Unfortunately, the reforming process generates impurities that may negatively affect FC performance. The effects of CO impurity have received most of the attention. However, there are other impurities that also may be detrimental to FC: operation. Here we present the effects of ammonia, hydrogen sulfide, methane and ethylene. Two structural domains of the membrane and electrode assembly (MEA) are usually affected by the presence of a harmful impurity. First, the impurity may decrease the ionic conductivity in the catalyst layer or in the bulk membrane. Second, the impurity may chemisorb onto the anode catalyst surface, suppressing the catalyst activity for H{sub 2} oxidation. Catalyst poisoning by CO is the best known example of this kind of effect. Fuel reforming processes [1] generally involve the reaction of a fuel source with air. The simultaneous presence of N{sub 2} and H{sub 2} may generate NH{sub 3} in concentrations of 30 to 90 ppm [1]. The effect of NH{sub 3} on performance depends on the impurity concentration and the time of anode exposure [2]. Higher concentrations result in more rapid performance decreases. If the cell is exposed to ammonia for about 1 hour and then returned to neat H{sub 2}, it will recover its original performance very slowly (about 12 hrs). This behavior is quite different from that of CO, which can be quickly purged from the anode with pure H{sub 2}, resulting in complete performance restoration within a few minutes. Longer exposure times (e.g. >15 hrs) to ammonia result in severe and irreversible losses in performance. It seems that replacement of H{sup +} ions by NH{sub 4}{sup +} ions, first within the anode catalyst layer and then in the membrane, is the primary reason for cell current losses. H{sub 2}S also adversely affects FC performance. Figure 1 depicts the current density changes in a FC exposed to both 1 and 3 ppm H{sub 2}S while operating at a constant voltage of 0.5 V. As expected, the greater the contamination level the faster the current density drops. Eventually in each case the cell becomes totally disabled. The effect H{sub 2}S appears to be cumulative, because even sub-ppm H{sub 2}S levels will decrease the FC performance if the exposure is long enough. We have recorded slow current droppings to about 20% of the initial value after exposure to concentrations of H{sub 2}S of 200 parts per billion (10{sup 9}) for 650 hours. Exposure to higher concentrations of H{sub 2}S may bring catastrophic consequences. We have exposed cell anodes to H{sub 2}S burps of the order of 8 ppm, and observed that the current at 0.5 V dropped from 1.1 to 0.3 A cm{sup -2} in just few minutes. Figure 2 shows the effect of H{sub 2}S on cell polarization. Curves b and c in this figure were recorded after 4 and 21 hours of exposure to 1 ppm H{sub 2}S, respectively, while keeping the cell at a constant voltage of 0.5 V. Regardless impurity concentration and running time, replacing the contaminated fuel stream with pure H{sub 2} does not allow any recovery as observed with CO poisoning. Cyclic voltammmetry (CV) indicates that H{sub 2}S chemisorbs very strongly onto Pt catalyst surface and high voltages are required for full cleansing of the H{sub 2}S-poisoned active sites. After full anode poisoning with H{sub 2}S (curve c), the electrode was subjected to CV (up to 1.4 V) and then the polarization curve d (with neat H{sub 2}) was recorded. The complete cell performance recovery is apparent from this curve. A more extended discussion on H{sub 2}S catalyst poisoning and cleaning will be presented. We also tested methane (0.5 % by vol.) and ethylene (50 ppm) as potential fuel impurities and we found no effects on performance.

Research Organization:
Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
Sponsoring Organization:
USDOE
OSTI ID:
975330
Report Number(s):
LA-UR-01-2675; TRN: US201008%%178
Resource Relation:
Conference: "Submitted to: Electrochemical Society Meeting, San Francisco, CA, September 2001."
Country of Publication:
United States
Language:
English