A mathematical model was developed to study the performance of methane (CH4) fed solid oxide fuel cells (SOFCs) considering the direct internal methane steam reforming (MSR) and water gas shift reaction (WGSR). An important feature of this model is that the effects of electrode structural parameters on both the exchange current density and gas diffusion coefficients are fully taken into consideration. The simulation results agreed well with literature data and thus validated the present model. Parametric analyses showed that all the overpotentials decreased with increasing temperature. This finding is different from previous analyses on hydrogen (H-2) fed SOFCs, in which the concentration overpotential is found slightly increasing with increasing temperature. This interesting phenomenon for CH4 fed SOFCs can be explained that the rate of MSR and WGSR increases with increasing temperature, leading to high rate of H-2 production inside the porous anode and a high molar ratio of H-2 to H2O- More simulation results were conducted to investigate the electrode's microstructural effects on the performance of CH4 fed SOFCs. It is found that increasing electrode porosity or pore size decreases concentration overpotential but increases activation overpotential of CH4 fed SOFcs. At low current densities, low porosity and pore size are desirable to reduce the electrode total overpotentials as concentration overpotential is insignificant compared with activation overpotential. At high current densities. the total overpotentials can be minimized at optimal porosities and pore sizes. In order to further improve the performance of CH4 fed SOFcs, advanced electrodes with porosity graded and pore size graded structures are evaluated. it was found that both porosity grading and pore size grading were effective to increase the SOFC working potential due to reduced concentration overpotentials. The present study provided better understanding on the coupled transport and chemical reactions (i.e. MSR and WGSR) at the porous electrodes. The model developed can be used to conduct more analyses for design optimizations. (C) 2008 Elsevier Ltd. All rights reserved.