Quantitative analysis of inclusion distributions in hot pressed silicon carbide

The main objective of this work was to investigate the relationship between microstructural defects and ballistic variability in hot pressed SiC. Two plates, 4” × 4” × 1” thick, from the same manufacturing lot, identified as samples A and B, were subjected to nominally identical high velocity impacts on the 4” × 4” face under which sample B performed more poorly than A. The exposed microstructural defects on fragments of the two samples were carefully characterized using scanning electron and optical microscopy. Two types of inclusion defects were identified: carbonaceous and aluminum-iron-oxide phases. While only small inclusions were found on polished cross sections, the rubble fracture surfaces were found to contain disproportionate numbers of large inclusions, strongly suggesting an influence on crack formation. A statistical function derived previously by Jayatilaka and Trustrum was successfully used to describe the defect populations. Sample A had more numerous, but smaller inclusions on its fragments than sample B. The exponent (n) of the distribution function can be related to the Weibull modulus (m) used to describe the strength variation of brittle materials by the relation m = 2n, a relationship that may be useful in efforts to represent defect distributions in micromechanical models. The data suggests that sample B has a higher probability of having a large inclusion on its fragments than sample A, which had more numerous smaller inclusions. Therefore, sample B would have a greater likelihood of crack nucleation and resulting failure at high strain rates from the distribution of large defects. The Pareto Principle, which suggests that 20% of the large inclusions in the tail of the distribution account for 80% of the total performance may be applicable.