Modelling cultivation and mechanical response of bacterial nanocellulose
File(s)
Author(s)
Koçkaldıran, Can
Type
Thesis
Abstract
Bacterial nanocellulose (BNC) is a material made of a network of nanocellulose fibres secreted by bacteria, most commonly Komagataeibacter xylinus. Both in its original wet pellicle form and dried nanopaper form, BNC receives considerable attention from the scientific community due to various desirable features such as high strength, porosity, and biocompatibility. Experimental studies were done about cultivation, refinement, and mechanical properties of this material, with a few simulation studies done to complement these. However, to this date, there was neither an agent-based simulation of BNC cultivation nor a mechanical simulation of 3-D BNC networks. This project is about constructing an agent-based simulation of BNC cultivation and using the resulting fibre networks for the mechanical simulation of wet BNC pellicles and compress-dried BNC nanopapers. The simulation provides a new method of analysing BNC materials and reveals new features of the material.
Cultivation simulation models oxygen and substrate diffusion as well as each individual cell’s movement, response to obstacles, entrapment by surrounding fibre network, growth, division, and metabolism. The simulation shows good agreement with experimental results. The simulation is employed to predict the sensitivity to various factors such as oxygen tension, pellicle vertical growth rate, cultivation vessel depth, and initial cell concentration.
Then, a model for chemotaxis of K. xylinus is added to the agent-based BNC cultivation simulation. Evolution simulations predict that chemotactic cells will outnumber non-chemotactic cells if they coexist in a cultivation.
Finally, mechanical simulations of BNC networks reveal the structure-property relationships for wet and compress-dried BNC networks. Fail stress of wet pellicle is shown to depend on crosslink density and average fibre cross-section area while fail strain is shown to depend on average segment length. Crosslink formation, hence strength of network, is shown to mainly depend on cells joining to/separating from previously laid BNC fibres.
Cultivation simulation models oxygen and substrate diffusion as well as each individual cell’s movement, response to obstacles, entrapment by surrounding fibre network, growth, division, and metabolism. The simulation shows good agreement with experimental results. The simulation is employed to predict the sensitivity to various factors such as oxygen tension, pellicle vertical growth rate, cultivation vessel depth, and initial cell concentration.
Then, a model for chemotaxis of K. xylinus is added to the agent-based BNC cultivation simulation. Evolution simulations predict that chemotactic cells will outnumber non-chemotactic cells if they coexist in a cultivation.
Finally, mechanical simulations of BNC networks reveal the structure-property relationships for wet and compress-dried BNC networks. Fail stress of wet pellicle is shown to depend on crosslink density and average fibre cross-section area while fail strain is shown to depend on average segment length. Crosslink formation, hence strength of network, is shown to mainly depend on cells joining to/separating from previously laid BNC fibres.
Version
Open Access
Date Issued
2025-10-01
Date Awarded
2026-02-01
Copyright Statement
Attribution-NonCommercial 4.0 International Licence (CC BY-NC)
License URL
Advisor
Tagarielli, Vito
Publisher Department
Department of Aeronautics
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)