Dataset collection underlying the thesis: "Virtual prototyping of wet granulation processes"
doi: 10.4121/e45fafed-843b-4db8-b095-7ea53383030b
Wet granulation processes are essential in various industries, such as pharmaceuticals, where they play a critical role in enhancing the flowability, compressibility, and homogeneity of powder blends -- properties essential for high-quality manufacturing. The (bio-)pharmaceutical industry is currently changing paradigms from traditional batch processing to continuous manufacturing, driven by the need for increased efficiency, flexibility, and product consistency. twin-screw wet granulation (TSWG) has emerged as a promising game changer, offering the potential for continuous processing with modular screw designs which can be adapted to specific formulations. However, process optimisation relies heavily on empirical adjustments and time-consuming experimental prototyping, which can be costly and inefficient. This doctoral thesis presents a comprehensive study on experiments and modelling of wet granulation processes, with a particular focus on TSWG. We aim to advance the field by developing a comprehensive multiscale modelling framework that integrates the discrete particle method (DPM), population balance (PB) models, and rapid prototyping, to optimise and predict wet granulation process behaviour, enabling virtual prototyping. The study highlights the significance of geometry and formulation dependencies on twin-screw granulator performance, identifies the limitations of existing models that rely on empirical adjustments and assumptions, and proposes the quadrature method of moments to model wet granulation processes. This is a collection of datasets which were published along with the thesis: "Virtual prototyping of wet granulation processes".
The first part of this thesis focuses on experimental TSWG investigations. Specific feed load, liquid-to-solid ratio, and screw configurations are identified as key parameters influencing particle size distribution, shape, and residence time. Utilising a full factorial design of experiments, this parametric study provides critical insights into the granulation kinetics and serves as a foundation for further modelling efforts.
In the second part, the thesis investigates DPM and PB methods individually. The DPM captures particle-scale dynamics, while the PB model is used to simulate the time evolution of particle size distributions. First, we introduce the quadrature method of moments for PB models of wet granulation processes. Second, we compare state-of-the-art calibration methods and propose different surrogate models to efficiently calibrate DPM simulations.
The final part of the thesis presents MercuryPBM, an open-source software for PB modelling of granular applications, which is designed to facilitate virtual prototyping of wet granulation processes. In the future, coupling DPM and PB methods allows for the prediction of granule properties under different screw geometries and material formulations, offering a more accurate and generalised model for TSWG. By incorporating experimental data with the developed advanced modelling techniques, MercuryPBM enables the virtual optimisation of process parameters and reduces the need for physical trials. This research not only contributes to a deeper understanding of TSWG but also provides a valuable tool for industrial applications, paving the way for more efficient and cost-effective wet granulation processes by virtual prototyping.
- 2024-12-03 first online, published, posted
- Virtual Prototyping of Particulate Processes (ViPr) – Design and Optimisation via Multiscale Modelling and Rapid Prototyping (grant code 16604) [more info...] Dutch Research Council
DATASETS
- [dataset] Data and code underlying the publication: Rapid Prototyping of a Twin-Screw Granulator for lab-scale research
- [software] Data to reproduce the paper: "Non-dimensionalization of quadrature method of moments for wet granulation"
- [software] Data to reproduce the paper: "Population balance modelling and reconstruction by quadrature method of moments for wet granulation"
- [dataset] Dataset as a basis for process modeling of twin-screw wet granulation: A parametric study of residence time distributions and granulation kinetics
