Addressing the global climate crisis necessitates innovative and sustainable approaches to carbon management that span capture, utilization, conversion, and long-term storage of CO2 (CCUS)–core tenets of a circular carbon economy. This work presents a computational framework grounded in multiscale modeling and physics-based machine learning to elucidate the fundamental mechanisms governing CO2 reactivity in condensed phase of liquid environments. Our research targets sustainable pathways for CO2 capture and utilization using green solvents, including biocompatible amino acid based ionic liquids (AAILs) and deep eutectic solvents (DESs). By integrating quantum and classical mechanics with enhanced sampling techniques and data-driven models, we provide molecular-level insights into proton transport phenomena, solvation structure and dynamics, and reaction energetics critical to CO2 capture reactions. These phenomena are contextualized within the broader framework of materials and reaction design for CCUS. Our approach bridges quantum chemical fidelity with micro-, meso-, and macroscale behavior, allowing predictive design of CO2-reactive solvents and functional materials that are efficient, non-toxic, and recyclable. The methodologies developed are generalizable to address grand challenges in sustainable chemical engineering, including the remediation of emerging contaminants and equitable access to carbon-negative technologies. This work advances the planetary boundaries discourse by offering practical routes to mitigate atmospheric CO2 concentrations while generating chemical value in a closed-loop manner–paving the way for resilient and regenerative systems in energy and sustainability.
