Dynamic compressive strength of alumina ceramics

Abstract

Dynamic (220–510 s−1) and quasi-static (0.001 s−1) compression experiments are conducted on alumina ceramics implemented with two types of tungsten carbide inserts, cylindrical and step-shaped. Split Hopkinson pressure bar (SHPB) tests with in-situ, high-speed optical imaging are adopted to capture the damage and failure of ceramic samples under dynamic compression. The compressive strength of alumina ceramic samples with step-shaped inserts is 15%–30% higher than that with cylindrical inserts commonly used in previous studies, under both dynamic and quasi-static loading. Damage occurs first at the two ends of ceramic samples with the cylindrical inserts, followed by edge fracture and splitting cracks penetrating the sample. However, damage is initiated in the sample region away from the sample ends for the step-shaped inserts, and oblique and secondary transverse cracks dominate the failure process. The different damage modes in the case of step-shaped inserts result in the delayed damage initiation and sample failure, and consequently high compressive strengths. Finite element modelling (FEM) of the SHPB tests provides strength and damage evolution features consistent with the experiment using the Johnson–Holmquist (JH-2) model. FEM reveals equivalent, tensile and shear stress concentrations at the two ends of samples with cylindrical inserts. The stress concentrations are responsible for the damage initiation and growth at the sample ends and the following splitting cracks, consistent with the high-speed images. In contrast, homogeneous stress distributions are achieved in the sample with the step-shaped inserts, ensuring simultaneous damage development across the sample. Overall, the step-shaped inserts in conjunction with cylindrical samples can yield reliable strength measurements for ceramics and ceramic-like materials.