Wood comes from trees, serves diverse purposes since a long time in human history such as building houses, planes, ships and making everyday furniture like tables and chairs. It’s an eco-friendly and lightweight material, however stiff and strong, making it an ideal choice for many people. At the micrometre scale, wood looks like a honeycomb material, with each cell wall made of a multi-layered composite strengthened by cellulose (a crystalline sugar) nanofibers known as microfibrils, which spirals at an angle (called MFA) to the cell axis. At the millimetre scale, its properties vary depending on three main directions (i.e., the tree axis, the growth rings radius, and tangential to the growth rings). It is also hygroscopic, meaning it can absorb or release moisture from its environment. All its properties naturally vary a lot from one sample to another due to the wide range of tree species, intra-tree variability, their genetic heritage and their growth conditions.
Despite the importance of this material, there are still unknowns regarding wood’s mechanical properties. For example, we don’t know very well how stiff a wood sample is when push or pulled in its transverse directions (radially or tangentially to the growth rings) or when twisted. Our understanding is limited for that material because we lack efficient and rapid tools to study these properties. However, understanding its stiffnesses, in other directions than the longitudinal one, and their origin is important for high-performance applications, such as parts for power generation, cars, planes, etc.
Our work aims to quickly measure as much properties as possible of a single wood sample from the CIRAD “xylotheque”[1], a wood collection that contains over 34.000 rectangular samples from 8.400 species from all around the world, all with the same dimensions (130 x 60 x 10 mm³), which cannot be modified as samples are precious. We do this by gently hitting the wood sample with a small hammer to make it vibrate and recording its vibrations using a small microphone[2]. Thus, by analysing the vibration of the sample, in bending and twisting in all directions, we can back-calculate its elastic properties without damaging it. Lastly, we study the mathematical relationships between the elastic properties of wood samples and their anatomical properties, focusing on density and the angle MFA rather than other anatomical features such as diameters, number of vessels, and wood rays. We used a wide density range of wood species (from the lighter, with density as low as around 0.2, to the heaviest, with density as high as around 1.3). This will allow us to predict the elastic properties of any wood parts just by measuring its anatomical properties, which are easy to measure.
Keywords: wood, orthotropic material, elastic properties, vibrational analysis, non-destructive testing
Authors
Alaa Al Fay
LMGC, University of Montpellier, CNRS, France
Olivier Arnould
LMGC, University of Montpellier, CNRS, France
Stéphane Corn
Université de Lorraine,LERMAB, France
Patrick Langbour
UR BioWooEB, CIRAD, France
Delphine Jullien
LMGC, University of Montpellier, CNRS, France
Login to download the PDF