A graph-based workflow for extracting grain-scale toughness from meso-scale experiments

We introduce a novel computational framework designed to allow an experimentalist extract the local toughness associated with crack growth along individual grain boundary from fracture experiments on standard specimens. The proposed relies ability of graph neural network perform high accuracy predictions micro-scale material toughness, utilizing limited size dataset that can be obtained experiments. analyze functionality different components framework, focusing minimum required for training following initial cycle synthetic dataset. demonstrate efficiency algorithm in accurately predicting (>68%) and global (>95%) scenarios where dependence microstructural parameters is substantially than one used merit arises its flexibility: it not specific mathematical description at simplicity: are rather standard. As such, enables study wide range microstructure properties investigation their influence toughness.