† Corresponding author. E-mail:
Dynamic failure and ejection characteristics of a periodic grooved Sn surface under unsupported shock loading are studied using a smoothed particle hydrodynamics method. An “Eiffel Tower” spatial structure is observed, which is composed of high-speed jet tip, high-density jet slug, longitudinal tensile sparse zone, and complex broken zone between grooves. It is very different from the spike–bubble structure under supported shocks, and has been validated by detonation loading experiments. In comparison with that under supported shocks at the same peak pressure, the high-speed ejecta decreases obviously, whereas the truncated location of ejecta moves towards the interior of the sample and the total mass of ejecta increases due to the vast existence of low-speed broken materials. The shock wave profile determines mainly the total ejection amount, while the variation of V-groove angle will significantly alter the distribution of middle- and high-speed ejecta, and the maximum ejecta velocity has a linear correlation with the groove angle.
Ejecta can be produced when a shock wave reflects from a metallic surface with imperfections where particulate fragments will be separated from the bulk and ejected outside the free surface at a higher velocity.[1,2] Ejecta formation is thought to be a limiting case of the Richtmyer–Meshkov instability (RMI) at a metal–vacuum (or gas) interface.[3,4] Because of potential damage and mixing impacted by these ejecta, this process has many important practical applications like pyrotechnics or inertial confinement fusion, and has motivated considerable research work worldwide.[1–5]
It is a general consensus that the physical mechanism for ejecta production is microjetting from surface imperfections (e.g., grooves, pits, or scratches) on the free surface of shocked metals.[1–5] The mass and velocity distribution of ejecta are relevant to various factors, such as shock strength and surface roughness. Zenller et al.[1,6] investigated the effects of shock breakout pressure, surface roughness, and groove angle on the ejecta production from Sn samples. Rességuirer et al.[2,7] studied the influence of various interface perturbation conditions on the laser shock-induced microjetting from the grooved metal surface. Chen et al.[8] confirmed experimentally the self-similar expansion evolution of ejecta on melted Sn sample subjected to high pressure loading. Chen et al.[9] and Shao et al.[10,11] simulated the micro-scale microjetting by molecular dynamics method, explained that the jetting fluid is the main mechanism of microjetting qualitatively, and obtained the change rules of microjetting characteristics in a wide range of pressure. Wang et al.[12,13] carried out a series of simulations by a smoothed particle hydrodynamics (SPH) method, and found that the ejection factor reaches its maximum at the half groove angle 45 degrees and reduces with the change of groove angle. They also found that the ejection process can be affected by the micro-structure of surface defect, besides the wavelength and depth of periodic grooves.
The previous focus of experimental and numerical research was the ejection behavior and influence factors under supported shocks, but unsupported shocks (shock waves followed by an unloading rarefaction wave) will be formed in high power laser and detonation loading conditions. When unsupported shocks reflect at the sample surface, the microspall will be formed near the free surface under the interaction of the incident unloading and free surface reflecting rarefaction waves, while the microjetting will be generated from the surface with interface defects owing to the RMI. Therefore, the mechanism of ejecta becomes more complicated due to the coupling between the microspall and microjetting. However, there is a little work about these issues and the mechanism of dynamic failure process is still not clear now.
In this paper, we simulate the ejecta production from typical periodic grooved Sn surface under unsupported shock loading, and investigate the dynamic failure process and ejecta characteristics. Furthermore, we analyze the influence of shock profile and groove angle on the total mass, spatial distribution characteristics of ejecta and the maximum velocity of jetting.
The geometric model in this paper is Sn that has periodic grooves on the surface, as shown in Fig.
Figure
We simulate the ejecta from the periodic grooved Sn surface using the in-house SPH program developed by Wang et al.[12,13] SPH is a Lagrangian particle code which uses no background mesh. In the SPH methodology,[14] smoothed particles are used as interpolation points to represent materials at discrete locations, so it can naturally obtain the history of particles and easily trace material interfaces. For a function f, its function value at a certain location can be expressed as summation interpolants over the neighbor particles using a smoothing kernel function W with the smoothing length h. The meshless nature of SPH methodology overcomes the difficulties due to large deformations. Various applications of SPH have been found in recent years, and it has been extended to high-explosive explosion, hypervelocity impact, and so on.[15,16]
The following equations about SPH are used to model the ejecta:
To obtain the deviatoric stress, we calculate the stress rates with the Jaumann stress rate tensor
The hydrostatic pressure part is calculated with metal polynomial equation of state.[14]
For the problem of material fracture, the stress components and pressure are set to zero when the maximum pull stress of the particle reaches the fracture strength and cannot withstand the negative pressure anymore. Metallic polynomial equation of state, Steinberg–Guinan constitutive model, and Lindeman melting law are used in the simulation. The material parameters of Sn are listed in Table
The ejecta at different times in θ = 120° and PSB = 33 GPa are shown in Fig.
The similarities and differences of ejecta are analyzed under supported and unsupported shocks at θ = 120°, PSB = 33 GPa, and t = 200 ns, as shown in Fig.
The mass and accumulated mass of ejecta along the impacting direction are analyzed under supported and unsupported shocks, as shown in Fig.
By tracking the move trail of Lagrangian particles, we analyze the material source of middle-high-speed ejecta. The distributions of material source of ejecta under unsupported and supported shocks are shown in Fig.
The simulation results of ejecta with different groove angles at PSB = 33 GPa and t = 410 ns are shown in Fig.
The mass and accumulated mass of ejecta along the impacting direction are also analyzed with different groove angles, shown in Fig.
Finally, the maximum ejecta velocities are extracted in Fig.
We have discussed the dynamic failure and ejection characteristics of a periodic grooved Sn surface under unsupported shock loading by a smoothed particle hydrodynamics method. An “Eiffel Tower” spatial structure is observed, which includes high-speed jet tip, high-density jet slug, longitudinal tensile sparse zone, and complex broken zone between grooves. It is quite different from the spike–bubble spatial structure under supported shocks and has been validated by detonation loading experiments. In contrast with the supported shock cases with similar peak pressure, the high-speed ejecta decreases under unsupported shocks and the truncated location of ejection moves towards the interior of Sn sample obviously, while the total mass of ejecta increases due to the vast existence of low-speed broken materials. The main mechanisms for these differences are the interaction of complex wave system and the deduction of shock pressure of unsupported shocks. Finally, we have studied the influence of interface defect angle on ejecta under unsupported shocks and found that the shock profile determines mainly the total ejection amount, while the variations of groove angle will significantly change the spatial distribution characteristics of middle- and high-speed ejecta. The maximum ejecta velocity shows a linear reduction with the increase of groove angle.
[1] | |
[2] | |
[3] | |
[4] | |
[5] | |
[6] | |
[7] | |
[8] | |
[9] | |
[10] | |
[11] | |
[12] | |
[13] | |
[14] | |
[15] | |
[16] | |
[17] |