Doctor of Philosophy
Chemical and Biochemical Engineering
Hrymak, Andrew N.
El Naggar, M. Hesham
Twin-screw extruders are continuous processing devices that have been used in many industries concerned with the processing of multiphase fibrous suspensions, such as the food and pulping industries. While these devices offer great flexibility due to their modularity and customizability, optimization for specific applications can be challenging due to the unavailability of reliable, global models with accurate predictive capabilities. Consequently, this work seeks to eliminate this knowledge gap with the development of a novel global, composite model capable of predicting key process outputs for suspension-processing, phase-separating, co-rotating, fully-intermeshing twin-screw extruders. Creation of the novel global algorithm involved the modification of existing plasticating twin-screw extruder modelling techniques. To account for two phase flow, the power-law model typically used to describe the rheological behaviour of molten polymer in plasticating extruders was replaced with the Herschel-Bulkley model, which has been used to describe the flow of fibrous suspensions in pipes. Terzaghi’s 1D consolidation equation was used to describe the phase separation taking place in the filtering sections of the extruder, and the absorption properties of the fibres in suspension were accounted for using a correlation developed for water-immersed woods. The extruder model was implemented as ‘TSESimFibre’ in MATLAB. A parametric study was carried out to assess the effect of operation parameters and screw configuration on the outcomes of the twin-screw extrusion system. The major performance parameter monitored was net liquid extraction rate from the extruder. The operation parameters assessed were the flow rate of wash liquid at the filter, and the dry fibre flow rate through the extruder. The screw configuration features investigated were the location of the reverse screw element, and the length of the filtering section. The simulations revealed that higher wash liquid flow rates and lower dry fibre flow rates were the best operating conditions to maximize net liquid extraction rates from the system. The simulations also revealed that longer filtering sections had smaller net liquid extraction flow rates, while the location of the reverse screw element had no effect.
Summary for Lay Audience
A novel, global simulator was developed to predict the system response of a solid/liquid separating twin-screw extruder used for the processing of wood fibre suspensions. To begin this work, literature on polymer-processing twin-screw extruders was explored to find equations relating predicted parameters (i.e. pressure, residence time, filling ratio …) to the extruder’s dimensions, operating conditions and processed material properties, and needed to be modified for the case of suspension-processing extruder applications. Modifications to those traditional modelling practices required the exploration of literature sources describing fibre behaviour across various disciplines. Techniques used to describe the rheological behaviour of flowing fibrous suspensions were gleaned from the materials literature, while that describing its consolidation behaviour was gleaned from the geotechnical engineering literature. The liquid absorption characteristics of the fibres in suspension was obtained from literature related to wood properties. Together, this information was used to modify existing extruder modelling procedures for two-phase, phase-separating twin-screw extruder applications. The novel model was validated using data obtained from Greenfield Global Inc. to assess the predictive capabilities of the simulator, and a parametric study was carried out to assess the effect of various extruder design features and operating conditions on pivotal model responses.
George, Keller J H, "Development of a Co-Rotating, Fully Intermeshing Twin-Screw Extruder Model for the Extrusion of Natural Fibre Suspensions" (2019). Electronic Thesis and Dissertation Repository. 6557.