Abstract
Scale-Up of an HME Process Supported by 3D SPH and 1D Mechanistic Modelling
Josip Matić*,**, Andreas Eitzlmayr*, Johannes Khinast*,**
*Institute for Process and Particle Engineering, Graz University of Technology, Inffeldgasse 13/III, 8010 Graz, Austria
**Research Centre Pharmaceutical Engineering GmbH, Inffeldgasse 13/III, 8010 Graz, Austria
Email for correspondence: khinast@tugraz.at
Introduction
The extrusion process, using twin-screw extruders (TSE) is well established in the polymer processing industry. Using the TSE, different continuous processes were developed, where the Hot-melt extrusion process (HME) is increasingly interesting for the pharmaceutical industry. The potential to increase the bioavailability of poorly soluble drugs, the steady product quality of a continuous process, the increased efficiency as well as a reduction of operational costs led to the specific interest of pharmaceutical industries in HME. In order to effectively predict the impact of screw configuration and process parameters on the HME process, an improved understanding of flow and mixing in individual screw elements as well as in the entire extrusion process is of high interest.
In our previous work, we showed that the Lagrangian based Smoothed Particle Hydrodynamics (SPH) simulation method provides significant benefits for extruder screw simulations. In contrast to Computational Fluid Dynamics (CFD), SPH inherently accounts for convective mixing and free-surface flows, allowing a detailed investigation of flow and mixing in fully filled and partially filled extruder screw elements. In order to provide correct boundary conditions for the complex screw geometries, a novel wall interaction method was developed.
Methodology
Due to the high computational expense of spatially resolved flow modeling, simplified modeling approaches are required for modeling of the entire extrusion processes. One-dimensional (1D) models consider the spatial dependencies along the screw axis and thus, provide the benefit of comparably low computational costs, suitable for the industrial use in design, optimization and scale-up of extruders. Following the idea of a continuous stirred tank reactor cascade with back-mixing, a 1D extruder model was developed in order to predict profiles of filling ratio, pressure, specific energy input, residence time distribution, material temperature and mixedness along the screw axis. Due to its strong simplifications, the 1D approach requires pre-computed data about flow and mixing in individual screw elements, which were obtained from the spatially resolved SPH model. This combination of the three-dimensional (3D) SPH approach for the detailed investigation of individual extruder screw elements and the 1D approach for the description of process variables along entire screw configurations results in a comprehensive tool, which increases the process understanding, supports design, optimization and scale-up, and reduces the need for experiments.
Results
An extruder scale-up scenario was analyzed based on proposed scaling laws for process parameters as throughput and screw speed which result in similar conditions at different length scales. Our results show the impact of the extruder dimensions on filling ratio, pressure, specific energy input, temperature distribution and residence time distribution.
Josip Matić*,**, Andreas Eitzlmayr*, Johannes Khinast*,**
*Institute for Process and Particle Engineering, Graz University of Technology, Inffeldgasse 13/III, 8010 Graz, Austria
**Research Centre Pharmaceutical Engineering GmbH, Inffeldgasse 13/III, 8010 Graz, Austria
Email for correspondence: khinast@tugraz.at
Introduction
The extrusion process, using twin-screw extruders (TSE) is well established in the polymer processing industry. Using the TSE, different continuous processes were developed, where the Hot-melt extrusion process (HME) is increasingly interesting for the pharmaceutical industry. The potential to increase the bioavailability of poorly soluble drugs, the steady product quality of a continuous process, the increased efficiency as well as a reduction of operational costs led to the specific interest of pharmaceutical industries in HME. In order to effectively predict the impact of screw configuration and process parameters on the HME process, an improved understanding of flow and mixing in individual screw elements as well as in the entire extrusion process is of high interest.
In our previous work, we showed that the Lagrangian based Smoothed Particle Hydrodynamics (SPH) simulation method provides significant benefits for extruder screw simulations. In contrast to Computational Fluid Dynamics (CFD), SPH inherently accounts for convective mixing and free-surface flows, allowing a detailed investigation of flow and mixing in fully filled and partially filled extruder screw elements. In order to provide correct boundary conditions for the complex screw geometries, a novel wall interaction method was developed.
Methodology
Due to the high computational expense of spatially resolved flow modeling, simplified modeling approaches are required for modeling of the entire extrusion processes. One-dimensional (1D) models consider the spatial dependencies along the screw axis and thus, provide the benefit of comparably low computational costs, suitable for the industrial use in design, optimization and scale-up of extruders. Following the idea of a continuous stirred tank reactor cascade with back-mixing, a 1D extruder model was developed in order to predict profiles of filling ratio, pressure, specific energy input, residence time distribution, material temperature and mixedness along the screw axis. Due to its strong simplifications, the 1D approach requires pre-computed data about flow and mixing in individual screw elements, which were obtained from the spatially resolved SPH model. This combination of the three-dimensional (3D) SPH approach for the detailed investigation of individual extruder screw elements and the 1D approach for the description of process variables along entire screw configurations results in a comprehensive tool, which increases the process understanding, supports design, optimization and scale-up, and reduces the need for experiments.
Results
An extruder scale-up scenario was analyzed based on proposed scaling laws for process parameters as throughput and screw speed which result in similar conditions at different length scales. Our results show the impact of the extruder dimensions on filling ratio, pressure, specific energy input, temperature distribution and residence time distribution.
| Original language | English |
|---|---|
| Publication status | Published - 2 Mar 2016 |
| Event | Jahrestreffen der ProcessNet-Fachgruppen Agglomerations- und Schüttguttechnik, Computational Fluid Dynamics und Mehrphasenströmungen - Bingen, Germany Duration: 29 Feb 2016 → 2 Mar 2016 http://processnet.org/AGG_MPH_CFD_2016.html |
Conference
| Conference | Jahrestreffen der ProcessNet-Fachgruppen Agglomerations- und Schüttguttechnik, Computational Fluid Dynamics und Mehrphasenströmungen |
|---|---|
| Abbreviated title | ProcessNet 2016 - AGG_MPH_CFD |
| Country/Territory | Germany |
| City | Bingen |
| Period | 29/02/16 → 2/03/16 |
| Internet address |