Moisture Transport in Wood Using a Multiscale Approach (Vocht transport in hout op basis van een multischaal benadering)

In this work the focus is on the multiscale behavior of wood. Where appl icable, the measurements are performed on the different structural levels to determi ne the wood characteristics. It is shown that the macroscopic material properties ar e originated in the wood microstructure. For example, the density is only dependent on the c ell wall density and the cell geometry. Therefore the material properties need to be correlat ed to the microstructure. One possibility is to separate the growth rings into a b i-layered representation, earlywood and latewood. However the continuous change in cell geometry from earlywood to latewood makes a representative separation difficult a nd therefore the analysis of the real cell geometry within the growth rings is essential. The further analysis of the material properties showed that the variatio n of the material properties within the wood needs to be taken into account. Where possibl e, this was already considered in this thesis. For example the density distribution within t he plank and also within the growth rings is analyzed. In the modeling part of the thesis, the different structural levels are analyzed from macroscopic representation over the bi-layered growth ring structure to the analysis of the cell geometry distribution within the growth rings. It is shown that a s imple unit cell approach already allows the investigation of the different cell types found in wo od. However, modeling of the real cell geometry shows that the consideration of the real cell geometry is important. A correlation between water vapor permeability and density is establishe d. The analysis of liquid uptake by X-ray radiography shows that preferenti al uptake occurs in the latewood layer. To model this behavior a phenomenological approach i s presented to derive liquid permeabilities for the bi-layered earlywood-latewood repre sentation. This model is already capable to model the preferential flow in the latewood layer, however due to the assumption of isotropic liquid transport, this approach could not comple tely describe all observed phenomena during liquid uptake. It is concluded that in modelin g of liquid transport, the cellular level needs to be taken into account.