Understanding the mechanistic behaviour of powder mixing with the use of DEM
The mixing of bulk solids is a fundamental unit operation in a wide range of manufacturing industries. Performing this operation effectively is essential to the quality of the final product but the mechanics of mixing in common mixing units such as paddle mixers is still not fully understood, resulting in poor product quality and economic losses to manufacturers. In this context fine granular materials such as powders are particularly challenging because their fine particle size distribution gives rise to a complex elasto-plastic-adhesive behaviour that produces unwanted segregation and agglomeration in mixing processes. Understanding the mechanics of mixing of such solids is experimentally challenging due to the opaque nature of the system making numerical methods such as Discrete Element analysis an indispensable tool of scientific investigation.
The current work provides a better understanding of the mechanics of mixing and segregation in a horizontally agitated paddle mixer via the Discrete Element Method, specifically focusing on the effects of particle size distribution, inter-particle adhesive forces and contact elasto-plasticity. A meso-scopic modelling approach is adopted using bi-sphered DEM elements to account for the non-sphericity of physical particles. Simulations with a very large number of particles are performed using GPU acceleration on an NVIDIA Quadro GV100 chipset to study the effect of particle size distribution on the mechanics of the system. The computational efficiency of this approach is explored by considering benchmarks and the results are included in this work. An innovative method for accessing and manipulating DEM data via the python scripting library EDEMPy is employed to calculate and visualise the spatial and temporal evolution of the mixture as well as important physical quantities such as the particle-particle contact forces and the kinetic energy distribution.