H2 Crossover

H2 crossover shows how much H2 crosses through the PEM to the cathode side. The US DOE target for the H2 crossover current density is less than 2 mA cm. For a pinhole-free PEM the H2 crossover current density is around 0.82, 0.66,0.33,0.19,0.13 and 0.09 mA cm for Nafion membranes with a thickness 

The H2 crossover rate can be determined similarly with the setup discussed in the ECSA section. One side is fed with humidified H2 and the other side with humidified N2. The N2 side is the working electrode and the H2 side is the counter electrode. The voltage of the N2 side can be set at around 0.5 V or scanned between 0.3 and 0.5 V. The H2 that crosses through the PEM to the N2 side will be quickly oxidized. The observed current density is the crossover current density. The amount of H2 can be calculated by Equation 4.19. 

---

As COR and OER occur simultaneously in the cathode, their kinetics are particularly important in evaluating carbon-support corrosion. The kinetics of OER is material-specific, dependent on catalyst composition and electrode fabrication.35,37 -39 A number of OER kinetics studies were done on Pt metal electrodes.37-39 However, there is a lack of OER kinetics data on electrodes made of Pt nano-particles dispersed on carbon supports. Figure 2 shows the measured OER current density with respect to the overpotential defined by Eq. (6).35 The 02 concentration was measured at the exit of a 50-cm2 cell using a gas chromatograph (GC). The 02 evolution rate (= 02 concentration x cathode flow rate) was then converted to the OER current density, assuming 4e /02 molecule. Diluted H2 (10%) and a thicker membrane (50 p,m) were used in the measurement to minimize H2 crossover from anode to cathode, because H2 would react with 02 evolved at the cathode and incur inaccuracy in the measured OER current density. Figure 2 indicates that the OER... 

Figure 1.19. Experimental OCV and theoretical OCV corrected for mixed potential and H2 crossover [22], (Adapted from Journal of Power Sources, 163, Zhang J, Tang Y, Song C, Zhang J, Wang H. PEM fuel cell open circuit voltage (OCV) in the temperature range of 23°C to 120°C, 532-7, 2006, with permission from Elsevier.)...

FIGURE 27.12 (See color insert following page S88.) O2 Permeability as a function of temperature and RH. Upper Umit (soUd line) defined by crossover losses (assuming no contribution from H2 crossover), lower limit (dotted line) defined by electrode ionomer film transport requirements, and data are for wet and dry Nafion 1100 EW-based membranes. (Reproduced from Gasteiger, H.A. and Mathias, M.F., in Proceedings of the Symposium on Proton Conducting Membrane Fuel Cells III, 2003. The Electrochemical Society of America. With permission from The Electrochemical Society, Inc.)... 

From AEa = [2.3RT/(omrF)]log(j/j°), the activation loss due to H2 crossover can be estimated. After H2 crosses through the PEM to reach the cathode catalyst layer, it will dissociate into protons and electrons according to Reaction... 

The relationship between the OCV and the H2 crossover current density is plotted in Figure 2.4. It can be seen that most of the OCV loss occurs in the extremely low crossover current density region (e.g., <10 mA cm ). 

It should be emphasized that the activation overpotential loss due to H2 crossover should be estimated using the ORR not the HOR. Do not be misled into thinking that it is the H2 that crosses through the PEM, and that it is the H2 that is oxidized at the cathode, and thus the HOR should be used to estimate the activation overpotenhal loss. 

Whether it is the H2 crossover current or the output current generated by a PEMFC, the impact on the activation overpotenhal loss is the same because the same reaction process occurs at the cathode. In actual tests, a PEMFC normally can give out a current density of 150 mA cm at aroimd 0.80 V, which is in reasonable agreement with the result shown in Table 2.4. 

When a fuel cell is at the open circuit, its voltage is also far lower than the thermodynamic voltage, mainly due to the crossover of H2 through the PEM from the anode to the cathode. At an H2 crossover current density of as low as 2 mA cm, the OCV drops from 1.18 V to around 1.04 V. But once the fuel cell is in operation, the voltage drop due to H2 crossover becomes unimportant. 

If we assume that the O2 solubility in Nafion is linearly proportional to its pressure, and the parameters for H2 are the same as those for O2, Table 3.15 lists the O2 and H2 crossover current densities at different pressures. 

Although the crossover current densities are not significant (for example, at O2/H2 pressures of 100/200 bars, the crossover current density is 21.2 mA cm 2), potential formation of explosive gas mixtures must be considered and avoided. The last two columns in Table 3.15 list the ratios of O2 or H2 crossover current density to the electrolysis current density (ECD) at 200 and 400 mA cm , respectively. For a mixture consisting of H2 and O2, the combustion range is around 4-94% vol. H2 (Wikipedia). Table 3.15 shows that the maximum H2 pressure within an electrolyzer cannot be higher than about 35 and 70 bars at 200 and 400 mA cm electrolysis current densities, respectively, in order for the H2 vol.% to be lower than 4%. 

Figure 4.13 shows the H2 crossover current density through an ePTFE-reinforced PEM with a thickness of around 18 jim. The H2 crossover current density is affected by its hydration level, and higher hydration levels lead to... 

The purpose of the temperature cycling test is to evaluate the impact of temperature change on the performance and the durability of the cell. The temperature at the higher end is the fuel cell temperature under operation, while the temperature at the lower end is the ambient temperature, which can be above or below 0°C. The fuel cell does not have to be running during the test. The impact of the temperature cycling can be estimated by periodically measuring the H2 crossover rate and the fuel cell performance. 

The theoretical OCV has the same value as the reversible eell potential. However, even when no current is drawn from a fuel cell, there is irreversible voltage loss, which means that the actual values of the OCV are always lower than the theoretically expected values. To date, a quantitative explanation for such OCV behavior has not been clear in the literature. One explanation attributes this behavior to H2 crossover and/or internal current, as described in the fuel cell book written by Larminie and Dicks [26]. A mixed potential [121-124] has also been widely used to interpret the lower OCV. The combined effects of fuel crossover, internal short, and parasitic oxidation reactions occurring at the cathode are the source of the difference between the measured open circuit cell voltage and the theoretical cell potential. Therefore, the actual OCV is expressed as... 

Figure 11.25. Illustration of calculated OCV drop vs. H2 crossover current density. The...

When the H2 erossover is measured by this type of eleetroehemieal method, the reaction at the eathode side is H2 2H + 2e. The H2 erossover rate is ealculated by n = i/2F. H2 erossover eurrent is an indieation of flie health of the membrane. It indirectly reflects the membrane decay rate, and is often used in accelerated testing. Figure 11.30 shows the H2 crossover current flnough a Hyflon membrane under H2/air, H2/N2, and N2/air, respectively [26], Under H2/air the crossover... 

The study indicated that the H2 crossover rate depends on the fuel cell operating temperature and backpressure, and the gas RH. 

Similar to H2 crossover in PEM fuel cells, NH3 can also cross from the anode (or cathode) side over to the cathode (or anode) side. Diffusion of NH3 in the membrane is rapid [85,89]. For instance, for a typical membrane thickness (Z) of 10 to 100 pm and a diffusivity of 10rrP/s, the estimated characteristic time constant for NH3 diffusion is of the order or 1... 

---