Improved Fiber Distribution Increases Thermal Insulation

Together with partners, we are investigating the effects of fiber distribution in wood-fiber based insulating materials with the aim of improving thermal insulation. The aim is to reduce the thermal conductivity to less than 35 W/K. The project is funded by the Federal Ministry of Education and Research (BMBF).

Insulation materials made from renewable raw materials are more sustainable and better for the climate than conventional insulation materials such as mineral wool and rigid foams, but they have one disadvantage: their thermal conductivity is in general higher. The distribution and orientation of the cellulose fibers are decisive for thermal insulation. We want to use the directional dependency of thermal transport and arrange differently oriented layers in such a way that the thermal insulation is optimized.

With porosity up to 90 percent, material density and thermal conductivity are proportional. However, this is no longer the case with highly porous materials such as the insulation boards examined; one reason could be thermal radiation, the influence of which has not yet been conclusively investigated.

Different Fibres Make Geometric Modelling Difficult

In addition to individual fibers, the insulation boards also contain fiber bundles of varying size, which makes their geometric modeling difficult. Our task is to process images and simulate the effective thermal conductivity of the wood fiber insulation boards. To do this, we analyze the wood fiber geometries using 3D image data, model the observed structures geometrically and finally simulate the heat transport.

The walls of the hollow cellulose fibers can only be mapped correctly at extremely high resolutions. This is only possible for tiny volumes, a maximum of one cubic centimeter.

We therefore model and simulate on two scales. The heat transport simulation on the fine scale is used to calibrate a simplified geometry model consisting of cylinders or beams. We ascribe »effective heat conduction properties« to them. From these simpler elements, we now generate virtual samples that are representative of the plate scale - i.e. the plate thickness. Finally, we optimize on this coarser scale by simulating heat conduction in ten cm thick insulation panels.
 

Extend Results to Other Highly Porous Materials

The methods developed in this project are not only intended to simulate and optimize heat transport in wood fibre insulation boards, but also to ensure improved simulation of heat transport in various highly porous materials. The two-scale approach to modeling and simulation as well as the image processing methods for thin-walled structures can be transferred to other material properties of wood-based materials such as the strength of furniture panels on the one hand and to other material classes such as cellular materials or highly porous battery and fuel cell materials on the other.