Characterising the linear viscoelastic behaviour of an injection moulding grade polypropylene polymer

The linear viscoelastic behaviour of an injection moulding grade polypropylene is studied using theoretical and computational methods. Polypropylene has a variety of engineering applications as a component. However, it commonly exhibits viscoelastic deformations. This paper analyses the creep and recovery responses of the BJ368MO polypropylene copolymer using the Burgers and generalised Maxwell models. Within the linear viscoelastic regime, an experimental creep strain at $20\ \text{MPa}$ 20 MPa is used to determine the rheological constants of the models. These constants (springs and dashpots) are determined using a nonlinear least-squares curve fitting of the experimental creep. Then they are used to predict the creep and recovery responses of the polymer at three different stresses, $10\ \text{MPa}$ 10 MPa , $12.5\ \text{MPa}$ 12.5 MPa and $15\ \text{MPa}$ 15 MPa . The experiments are made using tensile specimens designed according to the ASTM D638-14standard. The theoretical evaluations are made using the creep and recovery equations derived from their constitutive. Whereas COMSOL Multiphysics software is used during the finite element (FE) analyses. The results of the theoretical and FE calculations are verified using creep and recovery experiments. Based on the validation analyses, both viscoelastic models showed lower deviations from the experimental results when a computational approach is used. In addition, the viscoelastic models are compared by evaluating the residuals of the creep and recovery strain predictions. The theoretical analyses showed better predictions at $12.5\ \text{MPa}$ 12.5 MPa and $15\ \text{MPa}$ 15 MPa stresses when the generalised Maxwell model is used. However, the improvements are attributed to the recovery predictions. When FE is used, the Burgers model showed lower mean absolute percentage errors (MAPEs) in all creep and recovery predictions. The model has a minimum of 6.37% error at the $10\ \text{MPa}$ 10 MPa stress and a maximum of 8.23% error at the $15\ \text{MPa}$ 15 MPa . By comparison, the generalised Maxwell model showed a minimum of 9.24% error at $12.5\ \text{MPa}$ 12.5 MPa and a maximum of 12.8% error at $15\ \text{MPa}$ 15 MPa stresses. The novelty of this paper is on predicting the creep and recovery behaviour of the polymer using the FE and theoretical approaches in the linear viscoelastic regime. The findings suggest that the FE analyses using the Burgers viscoelastic material model provide better predictions, with all calculated errors falling below 10%.

Creep in oak material from the Vasa ship: verification of linear viscoelasticity and identification of stress thresholds

Creep deformation is a general problem for large wooden structures, and in particular for shipwrecks in museums. In this study, experimental creep data on the wooden cubic samples from the Vasa ship have been analysed to confirm the linearity of the viscoelastic response in the directions where creep was detectable (T and R directions). Isochronous stress–strain curves were derived for relevant uniaxial compressive stresses within reasonable time spans. These curves and the associated creep compliance values justify that it is reasonable to assume a linear viscoelastic behaviour within the tested ranges, given the high degree of general variability. Furthermore, the creep curves were fitted with a one-dimensional standard linear solid model, and although the rheological parameters show a fair amount of scatter, they are candidates as input parameters in a numerical model to predict creep deformations. The isochronous stress–strain relationships were used to define a creep threshold stress below which only negligible creep is expected. These thresholds ranges were 0.3–0.5 MPa in the R direction and 0.05–0.2 MPa in the T direction.