Tailor-Made Particles by Fluidized and Spouted Bed Spray Granulation: Opportunities and Recent Advancements

Spray granulation and coating are important process steps in the production of many high-value products. Besides the common fluidized beds, spouted beds can be used for gas-solid processes with liquid injection as the mushroom-shaped spout pattern offers advantages in terms of heat, mass and momentum transfer. In both apparatuses, spray nozzles inject a solution, suspension or melt into the bed of fluidized particles. The solidification of the liquid layer results in an onion-like growth of the granules. The main challenges are the formation of a homogeneous layer without cracks on one single particle and the homogeneous coating of the entire particle bed in one batch. In-line coating layer measurements are advantageous compared to off-line methods because real-time data are available and the process is not disturbed. Two different approaches for in-line monitoring of the coating progress in a laboratory spouted bed are presented: digital image analysis and optical coherence tomography (OCT). While the digital image analysis approach is able to detect qualitatively the coating homogeneity, the OCT method allows the quantification of layer thicknesses.

For industrial applications, the operation of spray granulation and coating processes is preferred on larger scales and in continuous mode of operation. In continuously operated spray granulation processes, horizontally constructed fluidized beds with rectangular cross sections are commonly used, which are divided by a variable number of plates (weirs) into several chambers for different functionalities (e.g. for granulation, coating, drying, cooling). Commonly, an external product processing, consisting of pneumatic conveying, screening, grinding of the oversize granules and recycling of grinded oversize and undersize particles into the fluidized bed granulator is applied. These can be modeled by multi-dimensional Population Balance Models (PBM) on the process unit scale. The resulting internal and external networks of solids process unit operations as well as gas, liquid and solid flows lead to a complex and dynamic process behavior, which can be captured in flowsheet simulations by combining the individual PBM’s. By adjustment of the process parameters the granule properties can be adjusted and optimized for example regarding the flowability or the morphology of the product. The spray rate and the drying temperature were found to have a major effect on the surface structure of granules.

Nowadays, computational methods offer great opportunities for understanding the details of processes on micro and macro scale. For spray granulation and coating processes, the CFD-DEM method is state of the art to simulate the interaction between fluid, gas and droplets. This method is very computationally expensive as both particle-particle and particle-fluid interactions are resolved. To cover the diversity of time scales present in industrial or even pilot scale plants, the recurrence CFD method (rCFD) offers remedy: by capturing the recurrent flow patterns present in the phase dynamics and extrapolating them in a physically meaningful manner, the computational effort is minimized and the numerically cheaper transport processes can be tracked on much larger time scales. The method’s applicability was validated by showing the accurate reproduction of instantaneous and average bed distributions and particle residence times in a spray zone. The speedup of rCFD over CFD-DEM for pure dynamics was 2100x for the laboratory spouted bed apparatus.