Enzymatic transformations of plant proteins require precise kinetic control due to the high cost of enzymes and the complexity of reaction mechanisms. In this study, an innovative approach is proposed to fit experimental enzymatic crosslinking data, to enable the modeling and simulation of plant protein crosslinking reactions. The transglutaminase-mediated crosslinking of rapeseed albumin (RA) was found to follow a distinctive evolution of crosslinked product formations that was independent of operating conditions, including pH, temperature, and enzyme concentration. Based on these trends, a unique crosslinking mechanism was identified and formalized through a system of crosslinking equations, describing the formation and consumption of individual crosslinked products. Under specific operating conditions, RA depletion kinetics exhibited a plateau behavior. The origins of this plateau were investigated and attributed to a combination of enzymatic deactivation and limited substrate accessibility. These mechanistic insights were incorporated into a kinetic model expressed as a system of differential equations and implemented in a fitting algorithm to describe experimental crosslinking kinetics across a broad range of pH and temperature conditions. The estimated kinetic constants were further regressed as a function of the enzyme-to-substrate ratio, showing strong correlations. The model accurately captured the evolution of RA monomers, dimers, trimers, and polydisperse products, enabling quantitative prediction of crosslinked protein product distributions under varying process conditions. This approach provides a robust framework for fitting experimental crosslinking data and establishes a predictive tool for simulation and optimization of enzymatic crosslinking processes in plant protein systems.
Hydrothermal carbonisation (HTC) of lignocellulose could improve its pyrolytic behaviour, but to date, few studies have evaluated the effect of HTC (150 degrees C and 200 degrees C) on fast pyrolysis. To fill this gap, this paper investigated how HTC affects the chemical composition of fir sawdust and, in turn, how these changes affect biooil yield and composition. HTC significantly alters the chemical properties of the solid material, affecting the content and acetylation degree of hemicellulose while removing inorganics (primarily K2O and CaO). Whereas the relative mass yield of pyrolysis products was not strongly affected by HTC, the pyrolytic behaviour was influenced by the HTC pre-treatment. HTC alters the rheological behaviour of the reacting material during fast pyrolysis, yielding hydrochar that displays an apparently melting and bed agglomeration, similarly to pure lignin, with ultimate fluidisation impairment. Fast pyrolysis products were characterised with an array of analytical techniques, allowing for the identification of changes in pyrolysis pathways and providing the tentative identification of novel holocellulose-derived oligomers in the aqueous phase portion of bio-oil. Beyond the obvious effects related to hemicellulose removal/deacetylation, results showed a meaningful change in biooil composition, with a strong increase in the share of anhydrosugars, which may be due to ash removal. Pyrolytic lignin was the least affected fraction, presenting only a slight reduction in molecular weight and a slight demethoxylation of its components. Results show that HTC at 150 degrees C/200 degrees C extracts or hydrolyses part or all of the hemicellulose and converts biomass into a material with different pyrolytic behaviour. In particular, hydrochar obtained at 200 degrees C behaves similarly to a heterogeneous mixture of pure lignin and cellulose, highlighting the approach's potential (increased selectivity toward anhydrosugars) and pitfalls (in-bed melting of the material).
This study investigates the shell-side hydrodynamics of a commercial cross-flow hollow fiber membrane contactor (HFMC), the Liqui-cel EXF 2.5x8 module. The shell-side flow was characterized through residence time distribution (RTD) measurements and computational fluid dynamics (CFD) simulations. The RTD experiments were conducted using a salt tracer technique. The numerical framework, based on the k-ε turbulence model, was validated against a Large Eddy Simulation (LES), demonstrating its capability to capture complex momentum and mass transport with sufficient precision. The results reveal that the dimensionless RTDs remain invariant over the investigated flow rate range, indicating that flow distribution is primarily governed by the internal geometry rather than inertial effects. Despite relying on theoretical estimates for the hydraulic resistance and dispersion within the fiber bundle, the CFD model exhibited strong agreement with experimental RTD profiles. Furthermore, a parallel-branch tanks-in-series model was successfully implemented to reproduce the global hydrodynamic behavior. The Peclet number associated with cross-flow flow through the fiber bundle exceeds 50, supporting the assumption of ideal plug flow in mass-transfer modeling. Overall, this work demonstrates that a physically grounded CFD approach provides a robust foundation for HFMC design optimization and industrial scaling.
