No abstract

No abstract

No abstract

The three-dimensional structure of a PEMFC catalyst layer (CL) was obtained using a dual beam Focused Ion Beam/Scanning Electron Microscope (FIB/SEM). Lower order statistical functions such as porosity and two point correlation functions calculated from the FIB/SEM data were used in the numerical reconstruction of a multi-phase CL domain. A 'carbon-sphere-based' initial seed structure, when optimized by simulated annealing, produced a structure for which the two point correlation function matched with the FIB/SEM data with very high fidelity. The reconstructed CL domain with its phase-resolved nano-structure was used to perform numerical simulation and predict effective transport properties. The coupled partial differential equations governing the reactive transport of charged and neutral species in the CL were discretized based on the finite volume method and solved implicitly using parallel computing. The simulated values of the effective oxygen and water vapor diffusivity, proton and electron conductivity, and thermal conductivity were in reasonable agreement with measured data reported in the literature. A parametric study was performed to investigate the impact of inaccessible pores-a feature that is not resolved experimentally. The simulations indicate that the properties most sensitive to the presence of inaccessible pores are the effective O 2 and H 2 O diffusivities and the effective proton conductivity.

Microscale flow simulation is considered in this paper for a microchannel flow geometry. Lattice Boltzmann Model (LBM) was used as the numerical method for flow simulation, in which an effective mean free path was used in relaxation time appeared in LBM. The effective mean-free-path makes it possible to investigate flow characteristics in transition flow regime, for which Knudsen number varies from 0.1 to 10. Such implementation does not change the computational efficiency of LBM significantly. Results are obtained for flow configuration in a long microchannel. The slip velocity was predicted in this flow configuration with good accuracy. Good correspondence with Direct Simulation Monte Carlo (DSMC) method was observed.