A novel finite element algorithm for predicting the elastic properties of wood fibers

Abstract

Wood fibers are industrially attractive low-cost natural materials that offer good mechanical properties. It is, however, extremely difficult to experimentally determine the elastic properties of single wood fibers due to the structural complexity and variability of basic properties. We propose a three-step finite element (FE)-modeling algorithm to predict the elastic constants of a single wood fiber. The model is based on calculating the elastic constants of the fiber in three consecutive length scales including nanostructure of cellulose microfibrils (25–30[Formula: see text]nm), ultrastructure in the fiber wall layers (2–3[Formula: see text][Formula: see text]m) and single wood fibers (30–40[Formula: see text][Formula: see text]m). The results for a given set of parameters are compared to previous studies with good agreement. The work exhibits its novelty through the model’s robustness and potential for industrial applications. It merely requires three essential inputs — chemical composition and bulk density of fiber and microfibril angle of [Formula: see text] wall layer, but is capable of predicting reasonably accurately the elastic constants of a wood fiber completely without any required model preprocessing or meshing like common commercial FE method software packages. Furthermore, the validated model is used to perform a parametric study. We have found that cellulose content has positive correlations with almost all the elastic parameters — relatively strong for [Formula: see text] and [Formula: see text], but weaker for [Formula: see text]. Lignin and hemicellulose have the greatest influence on [Formula: see text] and [Formula: see text]. The bulk density of fiber is shown to affect all elastic constants except the longitudinal elastic modulus.