On the role of the retained porosity on the shock response of additively manufactured high-performance steel: Experiments, constitutive model and finite-element predictions

Abstract

Experiments have shown that for quasi-static and moderate strain-rates (of the order of 102–103/s) the mechanical response of additively manufactured (AM) and traditionally processed high-strength steels is similar whereas the impact behavior is markedly different. In this paper, we reveal that the main reason for this difference is the retained porosity in the AM material. Fully-implicit finite element calculations are presented in which we simulate both the launching of the impact plate and the impact between the two plates. The constitutive model used is the elastic/plastic model for porous ductile materials with matrix displaying tension-compression asymmetry and Johnson-Cook hardening law that accounts for both strain-rate effects and plastic history. It is shown that even a very small initial porosity changes the wave front, decreases the Hugoniot while increasing the shock rise time, when compared to a void free material. Furthermore, quantitative comparisons between simulation results and plate impact data for both the AM and the wrought AF9628 steel are provided. The good agreement show that the model captures the impact response and illustrates the model capabilities to provide information on field variables that cannot be directly measured.Additive manufacturing (AM) of metals is rapidly advancing as a robust method for production of geometrically complex parts. To enhance understanding of material performance and open up additional application opportunities, dynamic characterization of newly printed alloys is required to validate their effectiveness. In this paper, we present results from plate impact testing of AF9628 steel, a newly developed high-strength low alloy martensitic steel for structural applications which require resistance to high-rate deformation. We put into evidence differences in the shock structure between the AM and the traditionally processed material. To gain understanding, we conduct fully-implicit finite element (FE) calculations in which we model both the launching of the impact plate and the impact between the two plates, respectively. An elastic/plastic damage model that accounts for the effects of the tension-compression asymmetry in plastic deformation and its influence on porosity evolution is used. The FE results reveal that even a very small amount of initial porosity leads to an increase in the shock rise time, explaining the observed trends. Furthermore, quantitative comparisons between simulation results and plate impact data for both the AM and the wrought AF9628 are provided. The good agreement show that the model captures the impact response and illustrates the model capabilities to provide information on field variables that cannot be directly measured.