Strain rate dependent material models for polymers

Abstract

The material behavior of polymers is known to be highly dependent on the rate of loading. However, lack of material data on different strain rates hinders the use of more accurate material models in Finite Element Method (FEM) simulations. In this thesis, material tests are conducted for five different polymers in both tension and compression by using the Split Hopkinson pressure bar system (SHPB) with three different strain rates. A servohydraulic testing machine is used to capture material behavior at low strain rates, and DMA-tests are conducted to investigate material transitions in the viscoelastic behavior. The achieved material data is then used for calibrating strain rate dependent material models and FEM simulations are performed, simulating a mobile phone drop case. The material tests confirm a clear dependency on the rate of loading for the tested polymers. Furthermore, the glass fiber (GF) reinforced polymers show brittle behavior as expected. This caused serious difficulties in the tensile high strain rate testing, and no reliable results were achieved for these two GF reinforced polymers. The test results from the compressive low strain rate tests were flawed due to the lack of proper instruments to precisely determine the test specimen strain levels. The objective was to use five different material models for the simulations: Elastic-Ideal plastic, Cowper-Symonds, Johnson-Cook, Parallel Rheological Framework and Mulliken-Boyce model. Unfortunately, simulation results for the latter two could not be achieved. Calibration of the material models was carried out using polynomial fits to the test data and MCalibration software. Simulation results show clear difference in strain levels between the models and suggest that the Johnson-Cook model provides the best results, when comparing to real life observations.