Polymeric materials are central to modern technologies, with applications ranging from energy to health and transportation. Although AI has made significant advances in materials discovery, the hierarchical structure of polymers across multiple length scales makes them inherently difficult to represent in a unified and physically meaningful way. Here we introduce HiPoly, a polymer-native AI framework that processes complete polymer descriptions through a three-level hierarchical graph architecture built on the G2RINS representation. HiPoly encodes stochastic inter-monomer connectivity, composition, and molecular weight directly within its architecture, using physically motivated design principles that mirror the multi-scale nature of polymeric systems. The framework establishes an end-to-end AI-driven workflow from experimental formulation data to property prediction, generative molecular design, and physics-based validation through molecular simulations, all unified by a single polymer representation. We demonstrate state-of-the-art prediction accuracy for thermophysical properties of multi-component polymer systems, with ablation studies confirming that each hierarchical design choice contributes independently to model performance. As an example, the generative design pathway is applied here to the discovery of sustainable alternatives to persistent fluorinated polymers, where it is possible to identify and independently validate PFAS-free candidates with target surface-energy properties. This work demonstrates how polymer-native AI can accelerate discovery by linking representation, prediction, and design across complex polymer chemistries.
Ge Sun, Gervasio Zaldivar, Yuan Tian et al.· 0 citations
Cooperative dynamics in soft materials such as colloidal suspensions, gels, and polymers stem from complex surface interactions and structural heterogeneities, driving behaviors such as yielding, failure, and avalanches. In X-ray photon correlation spectroscopy (XPCS), these dynamics manifest as localized decorrelation bursts in the two-time intensity correlation function, whose physical significance has remained difficult to quantify with traditional models that assume uniform and random motion. Here, we develop a deep-learning framework that detects and tracks such bursts as individual dynamical events. Combining theoretical validation with XPCS measurements of dense colloidal suspensions, we show that cooperative rearrangement regions (CRRs) are organized into temporally correlated hierarchies and cascades, with event durations and recurrence patterns that depart from homogeneous relaxation models. The results reveal a multiscale pathway for structural relaxation in dense colloids and demonstrate that intermittent features in XPCS encode physically meaningful cooperative dynamics. Our approach provides a route to quantifying avalanche-like and heterogeneous relaxation in soft, glassy, and disordered materials. We introduce an AI-powered framework that interprets intermittent bursts in two-time correlation functions as spatiotemporal "objects". Leveraging deep learning, it detects and tracks CRRs, extracting their occurrence times and durations to support detailed analysis. Validated through theory and experiment, this approach reveals the hierarchical and cascading nature of CRRs, offering new insights into relaxation dynamics in dense colloidal suspensions. It provides a robust tool for investigating complex behaviors in disordered systems.
Hong He, Yuan Tian, Heyi Liang et al.· Soft Matter· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
