Rational Design of Cleavable Polyethylene and Phenothiazine-Based Redox-Active Polymers Enabled by Simulation and Machine Learning

Developing materials for a sustainable future requires a holistic strategy that addresses current plastic pollution while anticipating future material needs. This dissertation explores three interconnected pathways using multiscale computational tools: deconstructing existing plastics, designing inherently recyclable polymers, and pioneering next-generation energy storage materials. First, we investigate catalytic hydrocracking of polyethylene (PE) under challenging conditions. By combining molecular dynamics simulations with the SAFT-γ Mie equation of state and physics-informed machine learning, we reveal how thermodynamic and diffusion properties evolve as PE melts transform into polymer–oligomer mixtures. These insights enable optimized reactor designs that enhance conversion and product yield. Second, we explore an alternative approach: embedding cleavable bonds directly into PE chains to enable controlled degradation. Through atomistic simulations, we assess ten bond chemistries and quantify their effects on melt density, chain mobility, crystallization, and solid-state structure. The resulting design guidelines balance processability with recyclability, offering a pathway toward truly circular polyolefins. Third, we address sustainable energy storage. Our computational analysis of redox-active polymers examines how anion choice, polymer–ion coordination, and backbone architecture influence ionic conductivity, swelling, and overall performance in all-organic batteries. This forward-looking work integrates circularity principles into battery materials to prevent future waste challenges. By connecting molecular-level understanding with predictive modeling, this dissertation charts a clear path from addressing current plastic waste to engineering high-performance, sustainable polymers for the future.