Validation of coarse grained DEM for simulating loose packing of cohesive powders

• Packing fraction of highly cohesive powder is largely affected by insertion methods. • It is important to consider collective states of particles to understand bulk powder behaviour. • Nearly identical clusters can be achieved by scaling initial fluctuation velocity with scale factor. • The SUP model can replicate the original packing fraction as low as 0.1. Coarse grain models for Discrete Element Method (DEM) have been attracting significant attention due to their potential to drastically reduce computational costs. However, their applicability in simulating loose packing behaviour of fine and cohesive powders remains unexplored. The purpose of this study is to validate the Scaled-Up Particle (SUP) model, which is a novel coarse grain model proposed in the authors’ previous work, for simulating loose packing behaviour of original cohesive powders. Two distinct particle insertion methods are employed to control the cluster configurations during free fall: random generation of individual particles with initial vertical velocity fluctuations and direct insertion of pre-formed particle clusters. The results show that the final packing structure are strongly influenced by the falling cluster configurations and the SUP model can reproduce a wide range of packing fractions tested in the original cubic particles, from 0.1 for highly cohesive particles to 0.8 for completely cohesionless particles. This highlights the importance of considering collective states of particles to understand bulk powder behaviour.