Numerical modelling of polyurethane used as core material in sacrificial cladding for blast mitigation

The increase of the number of intentional and accidental explosions nowadays, explains the interest in the development of suitable blast mitigation systems. One technique uses light weight sacrificial cladding which is composed of a polyurethane core and two thin metallic sheets. The polyurethane (PU) is used to absorb the blast wave energy generated by the explosion and to minimize the residual load applied to the main structure. For numerical prediction of the residual blast loads, one is confronted with the choice to make between different material models that are available in the finite element software, such as for example LS-DYNA. The present paper focuses on the comparison and reliability of the results of blast mitigation predictions with different material models for PU. Numerical analysis is performed with the following models in LS-DYNA: MAT026 (honeycomb), MAT057 (low density foam), MAT063 (crushable foam), MAT083 (Fu Chang foam), MAT126 (modified honeycomb) and MAT163 (modified crushable foam). These models are first used to simulate a quasi-static uniaxial compression test; the results are compared to experimental ones. Second, blast load conditions are simulated and strain rate effects are studied. The numerical results are compared to the ones obtained with analytical models, such as the one of Hanssen et al. A parametric study looks closer at the foam thickness and compared to the limited foam thickness given by the analytical approach. The paper shows that MAT026, MAT126 and MAT083 give the best agreement with the analytical results. The explanations presented in the paper show that care should be taken when material models have to be selected for numerical analysis of the performance of foam materials in blast mitigation applications.