Microstructure-Sensitive Enviro-Mechanical Response Characterization And Simulation In Sic/Sic Ceramic Matrix Composites

SiC/SiC continuous fibre reinforced ceramic matrix composites (CMCs) can exhibit a wide range of microstructure variability that arises from multiple unique processing routes (e.g., chemical vapour infiltration (CVI), polymer infiltration pyrolysis (PIP), melt infiltration (MI)). Service conditions of the targeted applications for these SiC/SiC CMCs produce a complex enviro-mechanical response. As life limiting damage initiation is often related to the weakest link or coupled extreme value microstructure attributes (e.g., pores, fibre clusters, secondary matrix phases) that conspire to produce the greatest driving force for damage initiation and propagation, this work is seeking to understand the relationship between local microstructure, microstructure variability, and the extreme value or anomalous attributes of the microstructure that have the most significant influence on the overall response. Here a set of automated tools have been developed to characterize the key microstructure attributes hypothesized to have the greatest influence on the damage response. Advanced segmentation and feature extraction algorithms are employed to separate out the various constituent phases, identify key attributes or features of interest and digitize the microstructure of the CMCs. Specific metrics are then defined to quantify the structural attributes and features of interest. Algorithms have also been developed to identify anomalous and/or extreme value attributes that are hypothesised to promote damage. Statistical volume elements (SVE) that sample the distributions of the microstructure attributes of interest are generated and discrete damage models along with environmental damage models are being developed to capture the microstructure-sensitive enviro-mechanical response. Experimental verification/validation is performed via specialized experimental techniques that identify areas in the material where damage initiates through acoustic emission detection (AE) and digital image correlation (DIC). Targeted 3D materials characterization of these damaged microstructure regions inform the stochastic models employed to describe the extreme value and/or anomalous microstructure attributes of interest.