Please cite this article in press as: Dreyschultze, C., et al. Influence of zone formation on stability of continuous fluidized bed layering granulation with external product classification. Particuology (2015), http://dx.doi.org/10.1016/j.partic.2015.02.004 ARTICLE IN PRESS G Model PARTIC-797; No. of Pages 7 Particuology xxx (2015) xxx–xxx Contents lists available at ScienceDirect Particuology j our na l ho me page: www.elsevier.com/locate/partic Influence of zone formation on stability of continuous fluidized bed layering granulation with external product classification C. Dreyschultze a , C. Neugebauer b , S. Palis a , A. Bück a , E. Tsotsas a , S. Heinrich c , A. Kienle a,b,∗ a Otto von Guericke University, Universitätsplatz 2, D-39106 Magdeburg, Germany b Max Planck Institute for Dynamics of Complex Technical Systems, Sandtorstraße 1, D-39106 Magdeburg, Germany c Hamburg University of Technology, Denickestraße 15, D-21073 Hamburg, Germany a r t i c l e i n f o Article history: Received 15 October 2014 Received in revised form 17 February 2015 Accepted 28 February 2015 Keywords: Fluidized bed Granulation Population balance Stability Bifurcation a b s t r a c t Continuous fluidized bed layering granulation with external product classification and a sieve-mill cycle can show instability in the form of self-sustained nonlinear oscillations of the particle size distribution. In the present study, the stability and bifurcation analysis of this process is presented. The underlying process models explicitly account for compartmentalization of the fluidized bed into a granulation and a drying zone, which is an important feature of many technical processes. Implications for plant operations are discussed with the help of stability diagrams as a function of zone size, residence time within different zones, the addition of external seeds and particular properties of the sieve-mill cycle. © 2015 Chinese Society of Particuology and Institute of Process Engineering, Chinese Academy of Sciences. Published by Elsevier B.V. All rights reserved. Introduction Fluidized bed layering granulation is used for the production of high quality particles from liquid suspensions or solutions in the chemical, pharmaceutical, and food industries (Kunii & Levenspiel, 1991; Mörl, Heinrich, & Peglow, 2007). During large scale appli- cations, these systems are operated as continuous processes with throughputs of up to several tons per hour. A characteristic flow- sheet is shown in Fig. 1. Typical systems employ a granulation chamber, where particles are fluidized by heated gas. Briefly, a solu- tion or suspension is sprayed into the chamber by a nozzle. Once inside the chamber, the droplets collide with the particles, spread, and then interact with the heated gas, which causes the liquid to evaporate and the solid to remain on the particle surface, result- ing in layer-wise growth in particle size. Particles are continuously discharged and classified into three fractions. The oversize fraction, ˙ n over , is ground in a mill and returned to the granulation cham- ber, which provides new nuclei, ˙ n mill , for granulation. In addition, external nuclei, ˙ n enuc , can be added. The intermediate fraction is the desired product, ˙ n prod , which is removed from the process. The ∗ Corresponding author at: Max Planck Institute for Dynamics of Complex Tech- nical Systems, Sandtorstraße 1, D-39106 Magdeburg, Germany. Tel.: +49 391 6110 369; fax: +49 391 6110 515. E-mail address: kienle@mpi-magdeburg.mpg.de (A. Kienle). undersized fraction, ˙ n fine , is returned directly into the granulation chamber. These processes are highly nonlinear and can give rise to insta- bility in the form of self-sustained oscillations, most prominently in the particle size distribution (Heinrich, Peglow, Ihlow, Henneberg, & Mörl, 2002; Schütte, Ruhs, Pelgrims, Klasen, & Kaiser, 1988). Similar patterns of behavior have been observed for other particle processes, including granulation processes with internal nucleation (Vreman, van Lare, & Hounslow, 2009) and continuous crystalliza- tion processes (Randolph, 1980; Randolph & Larson, 1988). These oscillations result in varying product properties, which are usually not acceptable. Oscillations can be avoided by suitable selection of operating and design parameters in the stable regime, or by means of stabilizing feedback control applied in the unstable regime (Bück, Palis, & Tsotsas, 2015; Chiu & Christofides, 1999; Christofides, 2002; Palis & Kienle, 2012, 2014; Vollmer & Raisch, 2001, 2002). The first strategy requires reliable prediction of parameter combinations leading to instability. Radichkov et al. (2006) generated a preliminary model based analysis of this cyclic behavior for the above process based on a simple model assuming a homogeneous granulation zone within the granulation chamber. This model assumes that the injected liq- uid is equally distributed on all particles, giving rise to a uniform growth that is proportional to the overall available particle sur- face according to Mörl et al. (Mörl, Mittelstraß, & Sachse, 1977; Mörl et al., 2007). Based on the assumptions of this model, which is http://dx.doi.org/10.1016/j.partic.2015.02.004 1674-2001/© 2015 Chinese Society of Particuology and Institute of Process Engineering, Chinese Academy of Sciences. Published by Elsevier B.V. All rights reserved.