(615h) Improving Drum Coating Processes by Computational Modeling and Experimental Evaluation

In the pharmaceutical industry, most of the produced tablets are coated, where the coating fulfills functional and non-functional purposes. A widely used technology is drum coating. Here, the tablets are placed in a rotating drum and the coating solution is introduced from above by means of a two-material nozzle. In order to increase the drying rate and to control the tablet temperature, a hot air stream is usually directed on the coating bed. The humid air flow is then extracted either through a perforated wall of the rotating drum or through an outlet duct. Although drum coating is a commonly used technology in the pharmaceutical industry, little theoretical and numerical modeling of internal gas flow and coating solution losses has been done so far. The choice of process parameters is often based on experience or trial-and-error practices.

The aim of this work is to obtain a better understanding of the thermo-dynamical factors affecting the quality of tablet coating process. For this purpose, modern CFD methods are used in combination with experimental measurements. The proposed numerical approach allows a deeper understanding of the reasons of operative problems (e.g. coating solution losses, filter plugging, too early evaporation and polymerization of spray droplets) and show potential for systems improvement. A central point is the application of the developed methods to industry-scale processes.

A core factor of the investigation is the CFD simulation of air flow inside the coater. The properties of the introduced air directly influence critical process parameters like tablet bed temperature or drying rate. An equally important factor is the investigation of the quality of the spray created by the two-material nozzle. For a widely used nozzle type, the distributions of droplet size, velocity and mass flow rate are investigated by means of a Phase-Doppler anemometry (PDA) system. The measurements are done at different locations inside the spray, and for varying values of atomization air, pattern air and liquid flow rate. Even if a uniform spray with good size distribution is generated, it is important to investigate how the path of the droplets is influenced by the air flow inside the drum coater. For this reason, an Euler-Lagrange DDM (Discrete Droplets Method) approach is adopted for the numerical analysis of the coating system. Models for the interaction between particles and tablet bed are also taken into account. The evaporation of computational parcels is furthermore modeled with a multicomponent approach.

The analysis of the interactions between air flow and spray, as well as between particles and tablet bed can finally help improving the system design. Analyzed values include the percentage of deposited film on the tablets in relation to the total introduced liquid and the uniformity of film thickness, especially important in the case of enteric coating, where porous films or films with thin spots can lead to rejected batches.