High-throughput dynamic experiments: The statistics of spall failure at ultra-high strain rates

Abstract

We describe a novel high-throughput experimental framework for dynamic loading based on an automated Laser-driven Microflyer Impact (LMI) platform. The framework is designed to rapidly explore the behavior of materials under dynamic loading, and we demonstrate this capability by performing an unprecedented number (173) of spall experiments to characterize the spall strength of a metal (polycrystalline pure copper). We examine pure polycrystalline copper with two different grain size distributions, one with a mean grain size well below 1.mum (referred to as the nanocrystalline copper) and one with a mean grain size  ∼ 8μm (referred to as the microcrystalline copper). The influence of microstructure and loading kinetics on the spall strength of metals remains poorly determined, largely due to the low throughput of conventional plate-impact experiments, motivating our study. We systematically vary the shock stress and (to a lesser degree) the tensile strain rate in our experiments, and examine the dependence of the spall strength on the prior shock stress, the tensile strain rate, and the grain size. The results provide clear evidence of a strong monotonic increase in spall strength with increasing tensile strain rate, and our microcrystalline copper results are largely consistent with the limited data available in the literature (often on material without a clearly defined grain size distribution). The influence of the prior shock stress on the spall strength is weaker than the strain rate effect, and appears (within our results) to depend also on the grain size distribution. Our results also demonstrate that our nanocrystalline copper has a significantly higher spall strength than its microcrystalline counterpart at these high strain rates. This is a reversal of the established trend that larger grain sizes can provide higher spall strengths when the grain size is sufficiently large, observed in some prior low-rate studies. Our extensive dataset provides validation data for microstructurally-aware spall models that predict an inverse grain-size dependence at high strain rates. This high-throughput LMI approach can also provide the datasets necessary for data-driven design of materials.