Catalyst layer modeling

The catalyst layer is platinum-based for both the anode and cathode. To promote hydrogen oxidation reaction and the anode uses either pure platinum metal catalyst or a supported platinum catalyst, typically on carbon or graphite for pure hydrogen feed steams. Oxygen reduction reaction at the cathode may use the platinum metal or the supported catalyst. For the HOR, the reaction is fast and can be described by a Bulter-Volmer kinetic expression. For the ORR, a Tafel equation is normal used.

The catalyst layer is very complex, because the electrochemical reactions take place there, and all the different types of phases exist. Thus, the membrane models and gas diffusion layer models must be used in the catalyst layer along with additional expression related to the electrochemical kinetics on the supported electro-catalyst particles. There are many approaches for studying the catalyst layer. An examination of the catalyst layer models reveals the fact that there are more cathode models than anode ones. Because the cathode has the slower reaction, it is where water is generated, and mass transfer effects are more significant. The anode can almost always be modeled as a simplified cathode model. The simplest approach is interface model, where a single equation is used. It is infinitely thin, and their structure can be neglected. This approach is used in complete fuel cell models where the emphasis of the model is not on the effects of catalyst layer.[13, 15, 16, 27, 53, 66, 67] If takes it as an interface, the results will be higher. The next set of models is the homogeneous models [64, 65, 68-70]. In the homogeneous model, it is assumed that the electrode active layer consists of a uniform, gas pore-free blend of proton conducting polymer and supported catalyst. The suitability of the homogeneous model for describing state-of-the-art PEM fuel cell cathode has been criticized.[28, 71, 72] Besides these models, the works of Wang et al. [73, 74]

treated the catalyst layer as an individual zone with various conservation equations employed in the modeling of transient study and various time constants for the transient transport phenomena were proposed. It predicts too poor oxygen permeability properties as compared to experimental results.

Experimental study and microscopic analysis [25, 71, 75] showed that the catalyst layer is porous and the reactants can transport through the catalyst layer in the gas phase. In order to account for this phenomenon, the so-called thin-film model and agglomerate model have been proposed. In the thin-film model [55, 76], the catalyst particles are covered by a polymer electrolyte film, and the gas pores only exist within the electrode. The thickness of the film is uniform and is very small in comparison to the pore size. The model of Bultel et al.[76]’s was developed for taking into account simultaneously the couple effects of diffusion and ohmic drops. They suggested that the local effects are mainly masked for oxygen reduction in acidic medium, and these effects are no more negligible for hydrogen oxidation.

In the agglomerate model [25, 68, 71, 72, 75, 77-79], the catalyst particles, electrolyte and gas pores form a homogeneous mixture. For the analysis, the effectiveness factor is used. For spherical agglomerate, an analytic expression is

( )

ϕ is the Thiele modulus[80] When the Thiele modulus is large, diffusion usually limits the overall rate of reaction. While it

is small, the surface reaction is usually rate-limiting. The results of the simple agglomerate models are helpful in trying to understand and optimize catalyst layer parameters such as catalyst loading and agglomerate size. The agglomerate model has many parameters that should be used to fit experimental data. Such as the agglomerate size and surface oxygen concentration. Several researchers had compared them to each other and experimental data. Broka and Ekdunge[71]

suggest that the agglomerate model is more accurate. Gloaguen et al.[77] also showed the agglomerate model is more suitable compared to the flooded thin-layer model in terms of describing the catalyst layer.

The National Research Council of Canada research group [81-89] had a series of investigation on the structures of catalyst layers with different types of agglomerate and optimizations of the cathode catalyst layer. However, their model only considers a single phase in catalyst layer or cathode side, while in reality multiphase of catalyst layers should be considered. For the flooding of the catalyst layer, there are various models have addressed. There are two main ways which depend on how the catalyst layer is modeled. If an agglomerate model is used and liquid water exists. A liquid film covering the membrane of the agglomerates can be assumed. Thus, the flooding of the catalyst layer is easily incorporated into the external mass transfer limitation.[29, 69] Because the low diffusivity and solubility of oxygen in water, only a very thin liquid film is needed to inhibit reaction. The thickness of the film used as a fitting parameter. The other approach is to use the two-phase modeling which described in the gas diffusion layer modeling. This involves calculating the liquid water saturation in the catalyst layer. The liquid water occupied the gas pore, and reduces the gas porosity. There are a few models that use this approach.[21, 33-35, 90, 91] Lin et al.[91]

developed a one-dimensional thin film-agglomerate model for the catalyst layer in the steady state.

They showed that the liquid water accumulation within the gas diffusion layer and catalyst layer had a significant impact on the cell performance. Lin and Nguyen [92] extended their one-dimensional model to a two-dimensional to account for the effect of the relative dimensions of the shoulders and channels on the cell performance. The effects of the in-plane liquid water permeability and electronic conductivity of the gas diffusionlayer on cell performance were also examined. It was foundthat more channels, smaller shoulder widths on the gas distributor,and

higher in-plane water permeability of the gas diffusion layercan enhance the transport of liquid water and oxygen, leadingto better cell performance.