(358a) Global System Analysis of Twin Screw Granulation Using Population Balance Modelling in gPROMS

Twin screw granulation (TSG) becomes increasingly popular as a method of wet granulation in the integrated part of continuous manufacturing for pharmaceutical industry. Despite many studies of TSG through the population balance model (PBM), the systematic analysis of the input parameters in PBM is still scarce. As a consequence, the influential parameters in PBM are yet fully identified and the holistic understanding of the granule growth in TSG remains elusive. This paper presents a global system analysis of the PBM operation for a ConsiGmaâ„¢ 25/1 continuous twin screw granulator. As a precursor to understand the influence of PBM parameters in the granule attributes, the modelling kernel formation in PBM is first defined in a compartmental way. The modelling formation consists of four rate process kernels, i.e. nucleation, layering, breakage and consolidation. Particularly, the recently developed breakage kernel is implemented in the software of gPROMS, which significantly improves the serviceability of breakage kernel for TSG. The feature of the breakage kernel is to take into account the powder feed number, dynamic yield strength and critical breakage size. The input parameters in PBM are identified and categorized whereas the reference values of all the parameters are specified. With the defined input value of PBM parameters, the comparative values of the testing parameters are also defined. With varying the comparative values of input parameters in a univariate way, the influence of input parameters on the granule size distribution (GSD) is investigated.

Apart from the rate process kernels, another two important process kernels i.e. liquid to solid ratio (LSR) and the average residence time in each compartment are also deployed for sensitivity study. Through the global sensitivity analysis of the input parameters in PBM kernels, the leverage of each parameter on the granule product attributes could be identified, which shed more insights on the validation of PBM model and model-based design of TSG products.