作者: Birgit Stute , Verena Krupp , Eric von Lieres
DOI: 10.1016/J.SIMPAT.2012.10.004
关键词: Mass transfer 、 Convection 、 Finite element method 、 SPHERES 、 Packed bed 、 Mechanics 、 Degrees of freedom (mechanics) 、 Work (thermodynamics) 、 Mathematical optimization 、 Computer science 、 Solver
摘要: Abstract Packed bed chromatography is commonly applied for the separation of large molecules in biopharmaceutical industry. A technical system typically composed a cylindrical column that filled with porous spheres. Particularly small columns, which are increasingly parallel experiments on lab robotic platforms, impacts inhomogeneous packing and wall effects performance can be quite significant. We hence study mass transfer by convection, diffusion adsorption three-dimensional sphere packings. Random packings externally generated imported into COMSOL where model equations easy to implement. However, functions automatic meshing iteratively solving resulting equation systems fail work default settings. have previously established semi-automated half-manual procedure rather less than 150 spheres works direct PARDISO solver. The present contribution addresses evaluation optimization iterative solvers provided given spatial geometries up ten million degrees freedom. presented results illustrate we solve 750 using memory computational time.