Bordered pits connect adjacent tracheid cells in softwoods and enable water transport between them. Knowledge of how large molecules, such as polysaccharides and enzymes, are transported through pits is important to understand the extraction process of valuable biopolymers from wood. The main mass transport mechanism for large dissolved molecules in wood is diffusion, and this is investigated through mathematical modeling in the lattice Boltzmann framework utilizing SEM images and 3D reconstruction of an actual bordered pit to compute an effective diffusion coefficient. Confocal laser scanning microscopy is used to find the unobstructed diffusion coefficients in a free aqueous solution using fluorescent diffusion probes of dextran. The effect of steam explosion on pit structure is explored through the use of a simplified model. The importance of different components of a bordered pit is investigated using simulation data, and results show that the most important structural features are the borders. Expressions for the effective diffusion coefficient as a function of the free diffusion coefficient are presented for a native and for a steam-exploded pit, respectively.

Diffusion of fluorescently labeled dextran of varying molecular weight in wood pretreated by steam explosion was studied with a confocal microscope. The steam explosion experiments were conducted at relatively mild conditions relevant for materials biorefinery at a pressure of 14 bars for 10 min. The method of fluorescence recovery after photobleaching (FRAP) was used to perform diffusion measurements locally in the wood microstructure. It was found that the FRAP methodology can be used to observe differences in the diffusion coefficient based on localization in the microstructure, i.e., earlywood, latewood, and cell wall. Microscopic changes due to steam explosion were seen to increase diffusion of the smaller 3-kDa dextran diffusion probe in the earlywood, while the latewood structure was not affected in any significant way. Macroscopic changes to the structure in the form of ruptures due to the steam explosion pretreatment were observed to increase the rate of diffusion for the larger 40-kDa dextran probe.

Research in the field of photochemistry, including photocatalysis and photoelectrocatalysis, has been revitalized due to the potential that photochemical reactions show in the sustainable production of chemicals. Therefore, there is a need for flexible photoreactor equipment that allows for the evaluation of the geometry, light wavelength, and intensity of the vessel, along with the fluid flow in various photochemical reactions. Light emitting diodes (LEDs) have narrow emission spectra and can be either pulsed or run continuously; being flexible, they can be arranged to fit the dimensions of various types of the reactor vessel, depending on the application. This study presents a 3D printed photoreactor with the ability to adjust distances easily and switch between high-power LED light sources. The reactor design utilizes customized printed circuit boards to mount varying numbers and types of LEDs, which enables multiple wavelengths to be used simultaneously. These LED modules, comprised of heat sinks and cooling fans, fulfill the higher heat dissipation requirements of high-power LEDs. The flexibility of the reactor design is useful for optimizing the reaction geometry, flow conditions, wavelength, and intensity of photochemical reactions on a small scale. The estimates for incident light intensity under five possible reactor configurations using ferrioxalate actinometry are reported so that comparisons with other photoreactors can be made. The performance of the photoreactor for differing vessel sizes and distances, in both the flow and batch modes, is given for a photochemical reaction on 2-benzyloxyphenol—a model substance for lignin and applicable in the production of biobased chemicals.

In high shear granulation it has been pointed out that there is a need for meso-scale resolution and coupling between flow field information and the evolution of particle properties. In this article we develop a modelling framework that compartmentalizes the high shear granulation process based on process relevant parameters both in time and space. It is built up by a coupled flow field and population balance solver and is used to resolve and analyze the effects of meso-scales on the evolution of particle properties. A Diosna high shear mixer is modelled with micro crystalline cellulose powder as the granulation material. The analysis of the flow field solution and compartmentalization allows for a resolution of the stress and collision peak at the impeller blades. Different compartmentalizations were done showing the importance of resolving the impeller region, both for only aggregating systems and systems with breakage. An investigation on the time evolution of the flow field depending on the changing particle properties was also done, indicating the importance of resolving meso scale phenomena in time as well as space.

Diffusion of large molecules throughout the porous microstructure of wood pretreated with steam explosion was investigated by using the lattice Boltzmann method for simulations. Wood samples were investigated with high-resolution X-ray tomography to effectively reconstruct an accurate geometry of the structural changes that ensue after pretreatment. Samples of approximately 1 mm 3 with voxel sizes from 0.5 to 1 μm were examined with X-ray imaging. These large volumes, relative to what reasonably can be simulated, were divided into sub-volumes and were further reconstructed into geometries suited for the LBM simulations. The transient development of the concentration was investigated, and the effective diffusion coefficient at steady state was computed. Diffusion rates were found to increase significantly in the transversal direction due to the steam explosion pretreatment. The increase was observed both in the time needed for solutes to diffuse throughout the pores and in the effective diffusion coefficient. A shorter diffusion pathway and a higher connectivity between pores were found for the pretreated samples, even though the porosity was similar and the pore size distribution narrower than the native sample. These results show that local mass transport depends not only on porosity but also, in a complex manner, on pore structure. Thus, a more detailed analysis of pore space structure using tomography data, in combination with simulations, enables a more general understanding of the diffusional process.

In this paper, multi-scale modeling was used to resolve diffusion in steam-exploded wood at tracheid scales including sub-micrometer features of bordered pits. Simulations were performed using the lattice Boltzmann method with high-resolution X-ray tomography image data as the input for the microstructure. The results show an effective method for utilizing a variable diffusion coefficient to implement two length scales. This circumvents the need to resolve the bordered pits in detail, which requires massive computing power. Instead, the effective diffusion coefficient for one bordered pit is used as input for this model. Results based on the present model are comparable to experimental data. This methodology can be extended to more structural features at the microscale of wood, such as latewood and the cell wall. Obtaining a map of different diffusion coefficients based on features and scale gives a better overall understanding of diffusion and the importance of mass transport with regard to the pretreatment of wood.