FLUIDIZED-BED MELT GRANULATION: POPULATION BALANCE MODELLING TO IDENTIFY THE AGGREGATION KERNEL

VILLA, MARTA; BERTIN, DIEGO; COTABARREN, IVANA; JULIANA PIÑA; BUCALÁ, VERÓNICA

Sheffield

Congreso; 7th International Granulation Workshop; 2015

Particles Product Group-University of Sheffield

ALCARACION: EL SIGUIENTE TRABAJO FUE ENVIADO EN OCTUBRE DE 2014 Y SE ENCUENTRA ACEPTADO PARA SU PRESENTACIÓN TIPO POSTER EN EL 7TH GW A REALIZARSE EL JULIO DE 2015. Within the operations that handle solids, granulation is considered as one of the most important processes. It allows producing granules with well-defined particle size distributions and shapes for many industries such as the pharmaceutical, agrochemical, detergent, food, etc. [1] [2]. Granulation processes are usually classified according to the binder nature as: wet, dry or melt. Nowadays, research in melt granulation has gained interest for many applications. Particularly for materials incompatible with water; the melt granulation avoids the use of solvents and the disadvantages associated with their recovery and final disposal, and minimizes the energy cost related to solvent evaporation [3] [4]. Among the melt granulation technologies, the fluidized-bed melt granulation is widely used. The final granules size distributions strongly depend on the operating conditions, the seeds size and binder properties. Modelling would allow improving the performance of the granulation process. However, the equation (population balance) that describes the particle size transformation as the granulation proceeds is quite complex involving kinetic parameters to describe the growth and breakage mechanisms and nucleation or elutration phenomena. Cotabarren et al. [2] and Veliz et al. [5 - 6] explored the effect of the operating variables on the process performance and particle properties for urea melt granulation carried out in a batch fluidized bed, identifying the operating growth regions that led to pure coating or combined growth by agglomeration and coating. Based on experimental data, (which are representative of relatively big seeds particles with respect to droplets size, high binder/seeds mass ratios, bottom spray configuration, and binder and seeds of the same chemical nature), in this work the kinetic parameters related to pure coating and combined mechanisms of coating and aggregation are established. As suggested by several authors [7 - 9], the aggregation kernel can be represented by two factors: one depending exclusively on the particle size (φ) and the other on the process operating conditions (β_0). Two different formulations are proposed: 1) φ is considered independent of particle size (i.e., φ=1) and 2) φ is represented by the EKE model [10]. For both cases, the kinetic parameter β_0 is adjusted for each experiment. Finally, a fitted mathematical expression for β_0 as a function of the binder flowrate, fluidization air velocity, bed temperature, atomization air flowrate and initial particle sizes, which allows predicting the experimental particles size distribution, is presented.