Many important products are made using fluidized bed reactors, where solid particles are suspended by a gas flow. This promotes highly efficient gas-particle contact, resulting in high heat transfer, high chemical reaction rates and high product yields. Multiscale modelling has proven to be indispensable in the design and optimisation of fluidized bed reactors. Most coarse-grained models assume that the solid particles are of spherical shape because this simplifies the treatment of gas-solid drag and particle collisions. However, many particles used in fluidized bed (bio)reactors are non-spherical. This means that anisotropic collisions, anisotropic gas-solid drag, effects of local particle alignment, and alignment by nearby internal and external walls all need to be taken into account.
I propose to pioneer a multiscale simulation methodology, backed up by validating in-house experiments, for prediction of structure formation in gas-solid flows of inelastic non-spherical particles. As a first step we focus on elongated particles. The multiscale approach consists of: 1) fully resolved simulations to obtain closures for translational and rotational gas drag tensors in crowded environments and near external and internal walls, 2) Discrete Particle Model simulations to validate the drag closures with matching experiments and to obtain statistics of angular and linear velocity changes due to inter-particle collisions between groups of particles, 3) a novel Lagrangian method based on stochastic multi-particle collisions. The collision propagation rules make maximum use of conservation laws and local symmetries of the particle configuration, orientation and deformation rates. The coarse-grained model is amenable to a parcel approach and can be coupled with heat and mass transfer models, allowing for simulation of industrial scale reactors with non-spherical particles.
Call for proposal
See other projects for this call
Funding SchemeERC-CG - ERC Consolidator Grants