Response and representation of ductile damage under varying shock loading conditions in tantalum

Abstract

The response of polycrystalline metals, which possess adequate mechanisms for plastic deformation under extreme loading conditions, is often accompanied by the formation of pores within the structure of the material. This large deformation process is broadly identified as progressive with nucleation, growth, coalescence, and failure the physical path taken over very short periods of time. These are well known to be complex processes strongly influenced by microstructure, loading path, and the loading profile, which remains a significant challenge to represent and predict numerically. In the current study, the influence of loading path on the damage evolution in high-purity tantalum is presented. Tantalum samples were shock loaded to three different peak shock stresses using both symmetric impact, and two different composite flyer plate configurations such that upon unloading the three samples displayed nearly identical "pull-back" signals as measured via rear-surface velocimetry. While the "pull-back" signals observed were found to be similar in magnitude, the sample loaded to the highest peak stress nucleated a connected field of ductile fracture which resulted in complete separation, while the two lower peak stresses resulted in incipient damage. The damage evolution in the "soft" recovered tantalum samples was quantified using optical metallography, electron-back-scatter diffraction, and tomography. These experiments are examined numerically through the use of a model for shock-induced porosity evolution during damage. The model is shown to describe the response of the tantalum reasonably well under strongly loaded conditions but less well in the nucleation dominated regime. Numerical results are also presented as a function of computational mesh density and discussed in the context of improved representation of the influence of material structure upon macro-scale models of ductile damage. V C 2016 AIP Publishing LLC.

Modeling and simulation framework for dynamic strain localization in elasto-viscoplastic metallic materials subject to large deformations

“…A general threedimensional implementation will be pursued in the future, in order to gain the ability to predict the spatial location and timing of shear band initiation more accurately. The initiation of such damage and failure processes is strongly linked with the local defect variability within the material, as shown by Bronkhorst et al (2016) in the context of void formation. Therefore, mean field representation of the material behavior will be inadequate in the future.…”

Microstructural effects on damage evolution in shocked copper polycrystals

“…simulations that model such conditions to provide input to the EVPFFT code. Here we use macro-and meso-scale FE simulations of the type reported by Bronkhorst et al [38,39] . As mentioned previously, the spall plane is a region of high tensile stress cause by wave interactions and it is the effect of this tensile pulse that is considered in this work.…”

“…In order to simulate microstructural effects of shock loading with EVPFFT, we need to obtain from FE simulations time-evolving strain-rate boundary conditions and Voce hardening parameters. For this, a FE simulation including a damage model described elsewhere [39] was run for polycrystalline copper under shock loading conditions, to the point in time where the model predicted that damage initiation occurred. The sample geometry and loading type in this simulation represented the experiment by Bingert et al [31] .…”

Thermodynamic theory of crystal plasticity: Formulation and application to polycrystal fcc copper

“…Despite more recent experimental work found in literature (Marchand and Duffy, 1988;Hartley et al, 1987;Duffy and Chi, 1992;Rittel et al, 2012Rittel et al, , 2017 thermometry remains challenging to perform accurately and more experimental work is required. This body of work strongly suggests that the Taylor-Quinney coefficient (Taylor and Quinney, 1934) is not a constant value as is commonly assumed (e.g., Bronkhorst et al, 2016) but can evolve significantly in magnitude with deformation. Recent work also suggests that the Taylor-Quinney coefficient differs in magnitude between slip and twinning processes (Kingstedt and Lloyd, 2019).…”