Characterization and evaluation of boron carbide for plate-impact conditions

Abstract

This article addresses the response of boron carbide (B4C) to high-velocity impact. The authors previously characterized this material in 1999, using the Johnson-Holmquist [AIP Conf. Proc. 309, 981 (1994)] (JH-2) model. Since then, there have been additional experimental data presented in the literature that better describe the hydrostatic pressure (including a phase change). In addition, a series of plate-impact experiments (one-dimensional, uniaxial strain) that used configurations that produced either a shock, a shock release, or a shock reshock was performed. These experiments provide material behavior regarding the damage, failed strength, and hydrostat for which previously there has been little or no data. Constitutive model constants were obtained for the Johnson-Holmquist-Beissel [J. Appl. Phys. 94, 1639 (2003)] model using some of these plate-impact experiments. Computations of all the experiments were performed and analyzed to better understand the material response. The analysis provided the following findings: (1) The material fails and loses strength when the Hugoniot elastic limit (HEL) is exceeded. (2) The material has significant strength after failure and gradually increases as the pressure increases. (3) The shear modulus does not degrade when the material fails (as has been postulated), but rather increases. (4) When the material is reloaded from an initial shocked (failed) state, the loading appears to be elastic, indicating the material is not on the yield surface after failure. To provide more insight into the behavior of B4C, the strength versus pressure response was compared to that of silicon carbide (SiC). The strength of SiC increases as the pressure increases beyond the HEL, probably due to pressure hardening or strain hardening. It appears that B4C does not experience any hardening effects and fails at the HEL. Although the HEL for B4C is higher than that of SiC, the hardening ability of SiC produces a similar maximum strength with more ductility. Another important issue with B4C is that there are significant differences between plate-impact data reported by different researchers. These differences are due to one or more of the following possibilities: errors in obtaining the test data, errors in analyzing the test data, a high degree of scatter due to failure of the material, and/or the different manufactured forms of B4C simply behave as different materials. These differences also make it difficult to validate the model (constants) determined from one set of data by applying it to other data reported in the literature. Lastly, the model was used to simulate ballistic impact experiments over a large range of impact velocities (1500–4500m∕s). The computed results overpredict the penetration at low velocities and underpredict the penetration at high velocities. Future work will address this inconsistency, but until this issue is resolved, the current model and constants should be used with caution when applied to ballistic impact and penetration.