Modelling and simulation of biorefinery processes Case study: Kraft pulping process

In the pursuit of a sustainable economy, the transition from petroleum-based to renewable feedstocks like lignocellulose demands a profound understanding of its inherent complex character during chemical processing. However, established modelling approaches for the kraft pulping process omit certain aspects of the hierarchical complexity of the feedstocks in use. Hence, these models must be revisited and gradually substituted by more rigorous approaches that capture the variability across multiple scales. They often overlook the intrinsic variability in feedstock properties, such as chemical composition. While some models exist, they typically focus on predicting average behaviour, e.g., the resulting pulp's kappa number, neglecting the distributed nature of initially presented chemical components within the wood and differences in its fragmentation kinetics. This work proposes novel modelling frameworks to address these limitations, introducing the concept of the distributed character for the chemical components within wood chips and investigating the resulting non-uniform delignification at a fibre level regarding individual fibre kappa number distribution. Furthermore, the frameworks incorporate the threedimensional properties of the anisotropic raw material structure and distinguish between different regions within the wood chip, e.g. early- and latewood regions. Resulting of adjustments regarding the kinetics of the well-established Purdue kraft pulping kinetics, a new kinetic model was developed, considering the heterogeneous nature of the lignin macromolecule with its diverse structures on a monolignol resolution using a graph and network structure, which allows a deeper understanding of the processes during fragmentation. Ultimately, this research aims to contribute to developing more efficient and targetoriented processes for producing bioenergy and biomaterials. Leveraging advanced modelling techniques and discussing the modelling results in the context of recent experimental findings offers insights that can help in decision-making and drive the transition towards a more sustainable future.