Paper to be presented at the Joint 19th airapt - 41th ehprg Int. conf. on HIGH PRESSURE SCIENCE AND TECHNOLOGY Bordeaux, July 7th - 11th 2003
Paper to be presented at the Joint 19th airapt - 41th ehprg Int. conf. on HIGH PRESSURE SCIENCE AND TECHNOLOGY
Bordeaux, July 7th - 11th 2003
K. Baumung a, G.I. Kanel b, and S.V. Razorenov c
aForschungszentrum Karlsruhe, P.O. Box 3640, 76021 Karlsruhe, Germany
bHigh Energy Density Research Centre of the Russian Academy of Sciences, IVTAN, Izhorskaya 13/19, Moscow 127412, RussiaRussian Academy of Sciences, Institute of Problems of Chemical Physics,
Chernogolovka 142432, Russia
We used a pulsed high-power proton beam of 50 ns fwhm duration to generate shock waves in solids by the direct effect of the beam on the sample or by impacting thin flyer foils launched to hypervelocities by the beam interaction. The power density distribution in the focal spot of the beam is bell-shaped with ~10 mm fwhm diameter and a peak value of 0.15 TW/cm2. With penetration depths of the 1.5 MeV protons in condensed matter of typically 5‑10 mg/cm2 a specific energy deposition of the order of 1MJ/g at specific power densities of 20 TW/g is achieved. The massive energy deposition drives a compression pulse with a nanosecond rise time and 20-30 ns decay period is driven into the adjacent cold material. With 50-100-µm-thick samples, reverberations of compression waves originating from the energy deposition zone and release waves starting from the rear free surface successively accelerate the residual cold part to velocities up to >10 km/s. The thickness of the compression wave in normal direction is 100-200 µm. This is small compared to the transverse fwhm of the pressure pulse of ~12 mm. Therefore, in a first approximation, we are dealing with planar shock waves with transversely varying pressure amplitudes. Thus, by placing the measuring section of an advanced line-imaging laser-Doppler velocimeter across the pressure gradient, we were able to perform measurements at different peak stresses of the compression wave in a single experiment.
This feature was particularly useful for determining the shock stress threshold which leads to melting of solids at isentropic unloading. By simultaneously recording the flyer acceleration and its effect on the target upon impact, we could determine the impact velocity which, at the rear free surface, leads to a loss of reflectivity indicative of melting.
The short load duration in the range of tens of nanoseconds allowed us to measure the dynamic tensile strength by the spall method at strain rates in the 106-107 s-1 range which is difficult to accomplish with conventional shock wave generators. The spall method is based on the tensile stresses which build up inside a sample when a shock wave is reflected from a free surface. The distance from the surface at which mechanical failure occurs depends on the pressure gradient in the rarefaction tail of the shock wave. For a given temporal profile of the wave this gradient depends on the peak stress. Utilizing the variation of the peak stress across the proton beam focus we were able to measure the tensile strength in 200-µm-thick functional metallic coatings on Inconel substrates simultaneously at various distances from the surface including the coating-bulk-interface and thus to measure the adhesive strength of the coating.
With aluminum single crystals we could observe that at the high strain rates we could achieve the tensile strength does practically not decrease with increasing temperature up to just 10 K below the melting temperature. These experiments also revealed that the dynamic yield stress increased by a factor of four when the temperature was increased from 20°C to 622°C. This is attributed to the dominating role of the phonon drag effect on dislocation motion in a single crystal.
Keywords: shock compression, spall strength, yield stress.