The performance of a proton exchange membrane fuel cell, PEMFC, is significantly affected by the rate of oxygen diffusion through the cathode catalyst layer, CCL. Continuum-scale modelling of PEMFCs requires knowledge of the effective oxygen diffusivity as a function of CCL porosity and its water saturation. To provide this functionality, we used three-dimensional pore-scale modelling to simulate the diffusion of oxygen under different liquid water saturations in CCLs having different porosity values. Solving the governing equations for immiscible two-phase flow, fluid distributions at different saturation levels were obtained. We show that the presence of liquid water initiates a hindering effect by decreasing the diffusive transport of oxygen. Oxygen diffusion, including dissolution of oxygen into the liquid water phase, was taken into account to calculate effective diffusivity values of the entire domain. The resulting effective diffusivity values showed good agreement with values reported in the literature, which are often based on quasi-empirical relationships. Utilizing a large number of simulation results, a correlation equation was developed for the effective diffusivity of oxygen as a function of porosity and liquid water saturation, which is appropriate to be used for macroscopic modelling of flow and transport in CCLs.

No abstract

No abstract

In this article, a previously developed and experimentally validated piston secondary motion model has been improved further numerically and applied to understand the detailed interactions between the piston skirt and the cylinder liner for various piston design parameters. The model considers the roughness of the surfaces and the topography of the skirt in both the axial (barrel profile) and circumferential (ovality) directions. Three modes of lubrication: hydro-dynamic, mixed, and boundary lubrication regimes have been considered and the skirt is partially flooded in most cases. Elastic deformation of the skirt is an essential part of the model. In this model, the piston dynamic behaviour and frictional and impact forces are predicted as functions of crank angle and are examined in detail. Parameters investigated include piston skirt profile, piston-to-liner clearance, surface roughness, and oil availability. The results show that some of these parameters have profound effects on the frictional and impact forces at the piston skirt/liner interface, and therefore, they have the potential to optimize the piston frictional power loss. Correlations and non-dimensional scaling laws are developed to illustrate the basic governing phenomena. These results aim to provide a set of quantitative design guidelines.

No abstract