Wood fiber boards are the most commonly used renewable insulation materials. However, their thermal conductivity is in general higher than that of conventional insulation materials (20-50K/W). Wood fiber insulation materials with porosities of over 95% achieve values below 38 W/K. We aim at further optimizing the board structure to achieve values below 35 W/K by a combination of mathematical image processing, stochastic geometry, numerical mathematics, and robust optimization.
Based on image data, simplified geometric structural models are developed.  They reflect essential geometric properties like orientation and size distributions as well as the distribution of solid mass in individual fibers, bundles, and bigger chunks while allowing for quick generation of large representative volume elements. Being a natural material, the microstructure of the wood fibers varies strongly, so that large volumes have to be analyzed and modelled to ensure representativeness. Moreover, the extremely high porosity makes it almost impossible to prepare sufficiently small samples for 3D imaging using computed tomography with sufficient resolution to preserve the thin walls of the hollow cellulose fibers.
Synchrotron radiation based computed tomography at ID19 of the ESRF allowed for completely preserved fiber walls thanks to sub-µ voxel sizes while keeping a rather large field of view thanks to ROI scanning and stitching.  Based on these images, we identify the contact regions of the fibers and separate individual fibers, bundles, and chunks, and finally fit a modified Altendorf-Jeulin fiber model. Numerical simulations of heat conduction in realizations of this fine scale model will be incorporated in a simplified geometry model on the next coarser scale. The cylinders forming the coarser model are calibrated to have a quasi-effective heat conduction. Virtual samples representative for the board thickness of about 10 cm can be generated from these cylinders. Heat conduction is numerically simulated in these virtual insulation boards, in order to ultimately optimize the structure on this scale.