Elastic Modulus as a Regime-Aware Computational Descriptor for Knowledge Extraction in Polymer/Clay Nanocomposites

In data-driven materials research, elastic modulus is often treated as a scalar target variable to be optimized. However, such an assumption obscures the regime-dependent nature of elastic responses in polymer nanocomposites. In this study, elastic modulus data of nanocomposite systems are analyzed using a regime-aware computational framework. Materials are first clustered based on the neat polymer elastic modulus, defining stiffness regimes that enable meaningful local comparison. Relative modulus changes are then represented through graph-based abstraction, allowing structural patterns to be extracted without black-box learning. The results show that elastic modulus exhibits amplified but unstable behavior in soft polymers, highly tunable yet chaotic responses in semi-soft systems, and increasingly constrained and predictable behavior in rigid polymers. Surface modification does not universally maximize modulus enhancement but consistently increases the organization and interpretability of elastic responses. The proposed approach demonstrates how elastic modulus can be transformed from heterogeneous numerical data into structured, explainable knowledge suitable for data-driven materials design.