Multiscale wood micromechanics and size effects study via nanoindentation

Wood as a material is a natural composite with a complex hierarchically arranged structure. All scale levels of wood structure contribute to its macroscopic mechanical properties. The nature of such characteristics and deformation modes differs radically at different scale levels. Wood macroscopic properties are well studied, and the relevant information can be easily found in the literature. However, the knowledge of the deformation mechanisms at the mesoscopic level corresponding to the cellular structure of early and late wood layers of annual growth rings is insufficient. It hinders building the comprehensive multiscale model of how wood mechanical properties are formed. This paper described the results of scanning of mechanical properties of softwood and hardwood samples, such as common pine, small-leaf lime, and pedunculate oak, by means of nanoindentation (NI). The NI technique allows varying the size of deformed region within a wide range by altering maximal load (P_max) applied to the indenter so that one can repeatedly and non-destructively test wood structural components at different scale levels on the same sample without changing the technique or equipment. It was discovered that the effective microhardness (H_eff) and Young's modulus (E_eff) decreased manifold with P_max growing from 0.2 to 2 000 mN. This drop in H_eff was observed when the locally deformed region grew, and resulting from P_max increase generally follows the rule similar to the Hall-Petch relation for yield stress, strength, and hardness initially established for metals and alloys, though obviously in those cases the underlying internal mechanisms are quite different. The nature and micromechanisms of such size effect (SE) in wood revealed using NI were discussed in this study. At P_max < 0.2 mN, the deformed area under the pyramidal Berckovich indenter was much smaller than the cell wall width. Hence, in this case, NI measured the internal mechanical properties of the cell wall material as long as free boundaries impact could be neglected. At P_max > 200 mN, the indentation encompassed several cells. The measured mechanical properties were significantly affected by bending deformation and buckling collapse of cell walls, reducing H_eff and E_eff substantially. At P_max ≈ 1–100 mN, an indenter interacted with different elements of the cell structure and capillary network, resulting in intermediate values of H_eff and E_eff. Abrupt changes in H_eff and E_eff at annual growth ring boundaries allow accurate measuring of rings width, while smoother and less pronounced changes within the rings allow identification of earlywood and latewood layers as well as any finer changes during vegetation season. The values of ring width measured using NI and standard optical method coincide with 2%−3% accuracy. The approaches and results presented in this study could improve the understanding of nature and mechanisms lying behind the micromechanical properties of wood, help to optimize the technologies of wood farming, subsequent reinforcement, and utilization, as well as to develop new highly informative techniques in dendrochronology and dendroclimatology.