Richtmyer-Meshkov instability for the analysis of materials at high strain rates

The mechanical resistance of a metal, as the strain rate increases, undergoes a significant increase in the range between 103 and 105 s−1 caused by the transition from a thermally activated mechanism to a viscous regime of the dislocations’ motion. In this regime, it is still possible, albeit with some difficulty, to obtain reliable high-strain-rate flow stress data from a small-scale Hopkinson bar test. Beyond this threshold, it is necessary to resort to impact compression experiments, which, however, generate very high pressures that do not allow for a straightforward estimation of the strain rate effect. The relatively low pressure obtained through Richtmyer-Meshkov instability (RMI) experiments has led to this technique receiving considerable attention in recent years. In RMI experiments, it has been demonstrated both theoretically and experimentally that the strain rate for a given impact is inversely proportional to the wavelength of the sinusoidal perturbations, provided the amplitude-to-wavelength ratio remains constant. Typically, in an experiment conducted with a gas gun, it is possible to obtain a strain rate greater than 107 s−1 with a perturbation wavelength of approximately 100μm. The material models used in the simulations can be validated by comparing both the deformation and the growth velocity of the spikes, which is a velocity measured experimentally using Photon Doppler Velocimetry (PDV). Spikes with a relatively small area can be a problem for both velocity measurement with PDV and post-mortem analysis. To overcome this limitation and modulate the strain rate, the application of RMI experiments was explored to investigate the mechanical response of annealed copper with perturbations larger than those reported in the literature. Numerical simulations were conducted using an implicit finite element code to demonstrate the feasibility of this study and to identify the limitations of the technique to extract accurate velocity data and fragments with manageable dimensions. In the simulation, a material model that includes an increase in resistance with varying strain rate was used. The numerical simulations indicated that the material points along the longitudinal axis represent various stages of a deformation history, taking into account the temperature effect. Subsequently, different geometries were selected from numerical simulations, and the test was conducted. The experimental results were compared with those obtained from the simulations.