Blood cell transport and aggregation using discrete ellipsoidal particles

A computational model is presented for efficient mesoscale simulation of the transport, collision and aggregation of blood cells, which can be applied to examine red blood cells (RBCs), leukocytes, or platelets in various types of blood flows in which the fluid length scale is substantially larger than the particle length scale. This method is intended to be intermediate between microscale models, which examine deformation and flow around a small number of individual blood cells, and more phenomenological continuum models. The computational model utilizes a particle approximation for the blood cells and introduces other physically-justifiable approximations in order to accommodate computations with large numbers of cells. For instance, the non-spherical RBC and platelet shape is incorporated into the model by use of ellipsoidal particles. A novel method based on particle level-surfaces is presented for rapid identification of particle collision. It is shown that receptor-ligand binding between the cells can be modeled under certain conditions using a formulation that is mathematically similar to van der Waals adhesion of particles, but in which the surface energy density is variable in time. The method is demonstrated to provide computations of the interaction and adhesion of over 13,000 red-blood-cell particles on an ordinary workstation. These computations exhibit formation of chain-like rouleaux aggregates, modification of rouleaux structure due to shear flow, and capture and/or breakup of colliding rouleaux. The model predictions are examined for rouleaux size distribution in channel flow in comparison to experimental data, as well as for the effect of RBC aggregation on margination of white blood cells and platelets in channel flows. (C) 2009 Elsevier Ltd. All rights reserved.