Packed beds of catalyst particles are normally described using models that contain a number of empirical parameters. The development of computer technology and CFD models makes it tempting to try to (1) fully simulate the flow in packed beds to obtain a more detailed understanding of the physical phenomena that take place in the bed, and (2) to use the CFD solutions to derive 'simple' correlations suitable for design purposes. In this paper it is shown that a commercial CFD code (CFX-5.3) can be used to predict, with an average error of about 10%, the pressure drop characteristics of packed beds of spheres that have a tube-to-particle-diameter ratio of 1.00 to 2.00. Packed beds with these unusually low tube-to-particle-diameter ratios can be used as unit cells in a novel type of structured catalytic reactor packing, proposed in this paper, that has very favorable pressure drop characteristics. The error of 10% in the pressure drop prediction by CFD is acceptable for design purposes. The CFD model is also able to predict local velocity profiles that were measured with LDA. The CFD results have been used to fit a simple two-parameter model that describes the experimental pressure drop data with an average error of about 20%. For a grid-independent CFD solution of laminar flow in a packed bed containing only 16 particles, already three million cells are required. However, it is anticipated that within five years from now the simulation of a packed bed containing a few hundred particles will be considered a;'standard' problem in terms of memory and calculation time requirements.