Viscoelastic behavior of softwood based on a multiscale computational homogenization

Abstract

In the present study, a numerical multiscale model is made to show how the hierarchical structure of softwood affect its macroscopic viscoelastic properties. The performance of the model is demonstrated for two softwood species — Norway spruce and Japanese cypress, whose creep behavior has been characterized experimentally. The results show that by using the same transversely isotropic viscous properties of the cell wall for both species, it is possible to predict creep deformation relatively close to experimental creep measurements for both species. Assuming that the variability is larger on the microstructural level (density, cell-wall geometry, microfibril angle, composition of wood tissues) than on the cell-wall level, it is possible to predict the macroscopic creep behavior based on the microstructural parameters alone. Such predictions can potentially save cost and time, since creep characterization in all material directions is demanding.