Determination qfeff ctive diffiisivity of a substrate it? biocatalyst particles is a key requirement in modeling heterogeneous reactions. The d ffusivity is mainly controlled by the molecular-diffi4sion characteristics oj'tl7e substrate, the tortuousness of the diftusion path within the biocatalyst, and the void,fi-action ol'particle volume available for cl ffusion. A general numerical model describing reactions in a biocatalyst particle following 7ero-ordcr, first-order, and Michaelis-Menten kinetics was developed. Finite volume method was used to discretize the nonlinear second-order differential equation, and tridiagonal matri.v allorithm was applied to solve the algebraic equation iteratively after discl-eti7ation. Computer codes ivere written for calculating the diffivsivity under specified boundary conditions. The numerical solution Qf the diffusion - reaction equations was validated against the e,vperimental data ' from literatures. The model was further calibrated with evperimental data obtained.fi-om.fungal pellet e-vperiments and then verified using additional data. The results show that the inverse methodology developed in this study was capable Ql"predicting d ffusivity in biocatalyst particles. Based on the predicted diffusivit ' v, oxygen consumption by an individual pellet was simulated, o.vygen consumptions by small versus large pellets were compared, and the effect of reaction rate on oxygen consumption was evaluated. (c) 2008 American Institute of Chemical Engineers.