Fluidized bed spray granulation is the preferred way of generating functional, high value granular products. By injecting a solution or suspension onto a fluidized particle bed, the particles grow layer by layer. The process conditions, specifically the drying potential, influence the state of the spray droplets when they impinge on the particles, for example the droplet viscosity and the relative velocity. Both quantities affect the spreading behavior of droplets on the surface, together with the crystallization occurring within the droplets themselves in the case of solutions, resulting in either porous or dense layers. As these phenomena are well accessible in CFD-DEM simulations, their combination can be a powerful tool in the design of granulators that produce tailor-made particles. A concept to realize this goal is presented: The surface structure of granulation products is measured for a variety of different process conditions in small-scale experiments. These experiments are replicated in silico using CFD-DEM, tracking the state of the droplets at impact, as well as the time for the evaporation of liquid layer on the particle surfaces. A mapping between these particle-scale quantities from the simulations and the resulting surface properties in experiments is derived, avoiding the direct dependency process conditions that may be very inhomogeneous in larger apparatuses. The concept is applied to evaluate the influence of granulator geometry (fluidized bed, spouted bed), nozzle configuration (top-spray nozzle, bottom-spray nozzle) and stabilizing internals (Wurster tube, draft plates) on the surface structure of granulation products.
