Linear modelling of biopolymer systems and related mechanical properties

A numerical method is proposed to assess the role of random microstructure on the effective Young’s modulus of a two-phase biopolymer composite material. An Ising model coupled to a Monte Carlo (MC) technique is used to generate virtual microstructures representing realistic starch–zein blends having random microstructure. The motivation here was to generate virtual microstructures that can be used in a numerical model to allow a continuous variation of both phase fraction and interface length. From the Pair Correlation Function (PCF), the minimum requirement for the Representative Volume Element (RVE) is established based on geometrical considerations. Finite element analysis allowed the prediction of the effective Young’s modulus as function of the phase ratio for the studied microstructures. The predicted trend is found close to that of Confocal Laser Scanning Microscopy (CLSM) microstructures of starch-based blends used as a case study. The comparison between the predicted results and the most popular analytical expressions points out that only the Hashin–Shrickman bounds are the most close bounds to the evolution of the effective Young’s modulus as function of second phase ratio.When implementing the intrinsic properties of starch and zein and considering virtual microstructures, analytical and numerical models exhibit the same trend. However, the comparison with the 3-p bending results suggests instead, a non-linear trend that can be inferred to the presence of imperfect starch–zein interface properties.