• Responsable de formations

• Sciences de l'ingénieur/Matériaux
  
• Sciences de l'ingénieur/Mécanique/Biomécanique
  
• Sciences de l'ingénieur/Mécanique/Mécanique des matériaux
  
• Sciences de l'ingénieur/Mécanique/Matériaux et structures en mécanique
  
• Sciences de l'ingénieur/Mécanique/Vibrations
  

PLA-flax non-woven composites are promising materials, coupling high performance and possible degradation at their end of life. To explore their ageing mechanisms during garden composting, microstructural investigations were carried out through scanning electron microscopy (SEM) and atomic force microscopy (AFM). We observe that flax fibres preferentially degrade 'inwards' from the edge to the core of the composite. In addition, progressive erosion of the cell walls occurs within the fibres themselves, 'outwards' from the central lumen to the periphery primary wall. This preferential degradation is reflected in the decrease in indentation modulus from around 23 GPa for fibres located in the preserved core of the composite to 3-4 GPa for the remaining outermost cell wall crowns located at the edge of the sample that is in contact with the compost. Ageing of the PLA matrix is less drastic with a relatively stable indentation modulus. Nevertheless, a change in the PLA morphology, a significant decrease in its roughness and increase of porosity, can be observed towards the edge of the sample, in comparison to the core. This work highlights the important role of intrinsic fibre porosity, called lumen, which is suspected to be a major variable of the compost ageing process, providing pathways of entry for moisture and microorganisms that are involved in cell wall degradation.

Fibres from annual plants are a sustainable alternative for glass fibres in composite manufacturing. However, both synthetic fibres, such as carbon fibres, and cellulosic fibres exhibit heterogeneities along their lengths, which appeared as localized morphological disorientations known as kink-bands. In plant fibres, kink-bands occur due to growing conditions under abiotic stress, during the retting stage and the fibre extraction process. Many studies have been conducted to identify the origin of such kink-bands and their impact on the mechanical behaviour of composites but their characteristics and fine ultrastructure remain unclear. Presence of cavities in transition zones was assessed by SEM and AFM and confirmed by low intensity SHG signal, especially when large size kink-bands are considered. Moreover, transverse indentation modulus is obtained by AFM in Peak force mode; no substantial differences were observed between the kink-band and defect-free regions with average values ranging from 6.2 to 7.3 GPa. Also, important MFA changes are 2 measured through SHG imaging, especially in large kink-bands with local misorientation up to 47°. Thus, this intense investigation of kink-band areas reveals ultrastructure heterogeneities and the presence of local defects.

Since the established correlations between mechanical properties of a piece of wood at the macroscopic scale and those of the cell wall at the submicron scale, techniques based on atomic force microscopy (AFM) have become widespread. In particular Peak Force tapping, allowing the differentiation of various layers, has become the new standard for wood cell wall's nanomechanical characterization. However, its use requires fully elastic indentation, a good knowledge of stiffness of the probe and assumes a perfect tip shape of known radius (sphere) or angle (cone). Those strong hypotheses can result in large approximations in the extracted parameters for complex, nanostructured, and stiff and viscous materials such as wood. In this work, we propose a reliable and complementary alternative based on AFM force-volume indentation by refining the Oliver and Pharr nanoindentation processing and calibration procedure for AFM cantilever and tip. The introduced areafunction calibration (AFC) method allows to considerably reduce these approximations and provides semi-quantitative measurements. No prior knowledge of the tip shape and cantilever stiffness are required and viscoplasticity is investigated through a qualitative index. Indentation parameters variations are shown to impact the resulting measurements, i.e., indentation modulus, viscoplasticity index, adhesion force and energy. AFC method, applied to map regions of tension wood, provides very stable mechanical parameters characteristic of each region, which makes this method of high interest for plant cell wall studies.

Nano-indentation instrumented tests are carried out at shallow depths on PAN-based and MPP-based carbon fibres. Indentation moduli are obtained by performing the tests at ten different measured orientations with respect to the fibre axis. They are used to identify the elastic constants of the fibres, assuming a transversely isotropic behaviour, by minimising a cost function between measured and estimated values. Inconstancies between the identified in-plane shear and transverse moduli and reported literature values are pointed out, and some drawbacks of the nano-indentation method are highlighted. An improved method taking into account the buckling mechanisms of crystallites at stake during the indentation process, and visible in the hysteretic behaviour of force-penetration nanoindentation curves, is proposed. It allows to identify values of elastic constants that are in accordance with literature values. These elastic properties of carbon fibres are in turn used to estimate the elastic properties of epoxy matrix composites containing these fibres. Very good agreement is found with experimentally available values of unidirectional ply properties. An excellent correlation between experiments and Finite Element Analyses of the indentation response of carbon fibres is eventually found.