EXPERIMENTAL PROCEDURES Two nominally closed-cell Al foams were investigated: (1) Alporas; relative density 0.15; cell size 23 mm; (2) Fraunhofer; relative density 0.24; cell size 1-2 mm. Cylindrical test samples (2.54 cm long x 2.36 cm diameter) were EDM-sectioned from blocks of each material. High strain rate (ε& = 400 s-1 to 2000 s-1) tests were conducted using a split Hopkinson pressure bar system, while low strain rate (10-5 s-1 to 1 s-1) tests were achieved using a servo-controlled hydraulic test machine. In some of the Alporas samples, 300 µm holes were drilled through the sample in order to produce interconnected cells. The spacing between the holes (4 mm) ensured that virtually all cells were thus penetrated. A high-speed camera was utilized to capture the deformation sequence for the Alporas foam during SHPB testing at high strain rates. Strain levels were determined within the sample using the recorded images and the Southwest Research Institute (SwRI) displacement mapping (DISMAP) system. This automated stereo-imaging technique, developed at SwRI, measures material deformation by mapping the displacements within the material [12]. Some high strain-rate compression tests were stopped at predetermined strain levels to allow evaluation of the deformed microstructures at a series of increasing strains; this was accomplished using metal spacers to limit the strain response. Samples were evaluated prior to testing, and after high strain-rate compression to approximately 3-4% and 9-10% strain. The 3-4% strain level was chosen since it represents a region on the stress-strain curve below the onset of the plateau region, while the 9-10% strain level is more representative of deformation throughout the plateau region. Following high strain-rate testing, samples of both foams were longitudinally sectioned by EDM for optical and scanning electron microscopy evaluation. Compressed samples were evaluated to determine the extent of cell wall deformation and the associated damage modes. RESULTS Both foams display a strain-rate dependence, examples of which are shown as stress-strain plots in Figures 1 and 2. For the Alporas, a significant increase in both yield and flow (plateau) stress (ε ≅ 5%) is evident at high strain rates; for example, at ε& = 700 s-1 (Figure 1). Under the same test conditions, the Fraunhofer material behaves differently (Figure 2). In this case, the material responds to the high strain-rate loading (ε& = 700 s-1, Figure 2) by experiencing a high stress pulse, which quickly drops to a plateau level that lies within the data band for quasistatic (10-5 s-1 ≤ ε& ≤ 1 s-1) experiments. Such peaks were never observed in any quasistatic tests, but were always present in high rate tests of Fraunhofer foam. For Alporas samples with EDM-interconnected cells, no high strain-rate sensitivity was observed (Figure 3). Moreover, comparison of Figure 3 with Figure 1 shows that apart from the typical high strain rate closed-cell test at ε& = 700 s-1, all of the data lie within essentially the same scatterband. The results of multiple tests at several strain rates are shown in Figures 4 and 5 for both foams, with strength plotted versus strain rate. For Alporas Al, the plateau strength for closed-cell samples clearly is strain-rate dependent at high loading rates (Figure 4), but it is essentially strain-rate independent for the drilled (interconnected cells) state. For the Fraunhofer material (Figure 5), strength enhancement at high loading rates is observed only if the height of the transient pulse is taken as the measure of strength. On the other hand, if the post-pulse (plateau) flow stress is plotted, it is found to be essentially strain-rate independent (Figure 5).
RkJQdWJsaXNoZXIy MjM0NDE=