No abstract

This work intends to decipher the role of hydrophilic fillers, wheat straw fibers (WSF), on the water vapor transfer properties (sorption and diffusion) of biocomposites based on poly(3-hydroxybutyrate-co-3-hydroxyvalerate), (PHBV) as matrix. Transfer in biocomposites, measured using dynamical vapor sorption measurement, is correlated to the transfer properties of each individual constituent and to the specific structural arrangements induced by the presence of particles inside the matrix. Increasing amounts of WSF particles in the PHBV matrix lead to an increase of the water vapor sorption (WVS) of the resulting composites. This is attributed to the high sorption of hydrophilic WSF as compared to that of the neat PHBV matrix. Water vapor diffusion in composites (around 0.13 × 10−11 m2 s−1 at 20°C whatever the filler content) is always lower than in the neat matrix (0.26 × 10−11 m2 s−1) although wheat straw displays high diffusivity values (1.84 × 10−11 m2 s−1). Such unexpected behavior is related to (1) changes of structure and properties of the WSF particle once embedded in the PHBV matrix, (2) changes in the polymer matrix structure and properties in contact with fibers, and also (3) to the representativeness of the filler diffusivity, which is difficult to appraise

A three-dimensional model structure that allows considering interphase layer around permeable inclusions is developed to predict water vapor permeability in composite materials made of a matrix Poly(3-HydroxyButyrate-co-3-HydroxyValerate) (PHBV) including Wheat Straw Fiber (WSF) particles. About 500 two-phase structures corresponding to composites of different particles volume fractions (5.14−11.4−19.52 % v/v) generated using experimental particles’ size distribution have permitted to capture all the variability of the experimental material. These structures have served as a basis to create three-phase structures including interphase zone of altered polymer property surrounding each particle. Finite Element Method (FEM) applied on these structures has permitted to calculate the relative permeability (ratio between composite and neat matrix permeability P/Pm). The numerical results of the two-phase model are consistent with the experimental data for volume fraction lower than 11.4 %v/v but the large upturn of the experimental relative permeability for highest volume fraction is not well represented by the two-phase model. Among hypothesis made to explain model’s deviation, the presence of an interphase with its own transfer properties is numerically tested: numerical exploration made with the three-phase model proves that an interphase of 5 µm thick, with diffusivity of Di≥1×10−10 m2·s−1, would explain the large upturn of permeability at high volume fraction.