Using Uncertainty Quantification to Guide the Design of Ceramics Subjected to Hypervelocity Impact

Engineering ceramics are often employed as the material-of-choice for protection against hypervelocity impacts. Dynamic failure of such materials is a complex process involving multiple failure mechanisms operating at several time- and length-scales. Failure can be through cracking-induced damage and degradation of elastic properties, as well as amorphization and/or lattice plasticity in zones under high pressures. In addition, the material may be comminuted in regions of high-deformation leading to granular flow-type behaviour. One approach to simulating this process has been through physics-informed models. While these models capture the mechanisms and are applicable to a wide range of conditions, they tend to be expensive, often involving many parameters. Phenomenological models, on the other hand, offer the advantages of simplicity and computational efficiency, but require fitting parameters that may not be appropriate for a given impact configuration. In this study we model dynamic failure using a ‘lower-scale’ multi-mechanism model as well as a ‘higher-scale’ phenomenological model and quantify uncertainty as it propagates from the lower to the higher scale. We perform sensitivity analysis over the large parameter space of the lower-scale model and provide failure probabilities under one (or more) hyper-velocity impact scenarios using the higher-scale model. We envisage applying the insights gained from this study to inform the design of engineering ceramics for such applications.