Hofegger, J., Königsberger, M., Pech, S., & Füssl, J. (2026). Effect of cell wall architecture on stiffness and microstresses in softwoods: microstructure generation and finite element modeling. Materials & Design, 270, Article 116870. https://doi.org/10.1016/j.matdes.2026.116870
The geometrical features of softwood’s cellular microstructure determine its mechanical clearwood properties, particularly the transversal properties perpendicular to the grain. These structure-property links are usually modeled using analytical mean-field homogenization schemes or unit cell models, which, however, cannot capture the stochastic geometric variations inherent in naturally grown softwood. To address this gap, we first developed a fully parameterizable softwood microstructure generator for the transverse plane. Distortion maps with Gaussian smoothing are used to transform an initially regular grid into stochastically varying grids, from which realistic wood cell architectures are reconstructed using Bézier curves. Next, hundreds of these cellular geometries were translated into finite element meshes and modeled using generalized periodic boundary conditions, applicable to the generated non-periodic clearwood structures. Systematic sensitivity studies revealed that irregularities in cell geometry moderately reduce the effective stiffness but strongly increase microstress concentrations in both the middle lamella and the secondary cell wall, compared to a structure with perfect geometry. Moreover, different stochastic realizations of the same microstructural parameter set lead to noticeable variations in mechanical behavior, highlighting the importance of stochastic geometry for microstructure-based modeling of softwood.
en
Project title:
AI-driven multi-scale methodology to develop Transparent Wood as sustainable functional material: 101138191 (European Commission) Advanced Computational Design: F 77 (FWF - Österr. Wissenschaftsfonds)
-
Research Areas:
Biological and Bioactive Materials: 30% Modeling and Simulation: 40% Computational Materials Science: 30%