Chinmaya Sahoo
Macroscopic mechanical behavior of epoxy nanocomposite adhesives is governed by composition-dependent evolution of elasticity, plasticity, and fracture. In this work, we in- vestigate epoxy nanocomposites modified with either stiff nanoclay or soft block-copolymer inclusions through uniaxial tensile experiments with full-field digital image correlation, which yield the Young’s modulus, maximum stress, ultimate stress, and ultimate strain of each specimen. These measured properties serve as targets for a Bayesian data-driven framework that combines composition, processing variables, and quantitative fracture-surface descrip- tors. Nanoclay inclusions increase stiffness and yield strength while suppressing plastic deformation and promoting brittle failure, whereas block-copolymer inclusions reduce both elastic and plastic resistance while accelerating the ductile-to-brittle transition. Specimens of identical nominal composition nevertheless exhibit measurable scatter in their mechani- cal properties, reflecting processing-induced defects that composition alone cannot describe. Incorporating fracture-surface morphology markedly improves predictions of the measured properties and their associated uncertainties, demonstrating that post-failure morphology encodes the realized mechanical state beyond nominal composition and processing condi- tions.