Structure-Property Relationships in Mixed Ionic-Electronic Conductors via Machine Learning-Enhanced Multiscale Modeling

Polymers possessing electronic and ionic coupled functionalities offer unique solutions for biomedical sensors, neuromorphic computing, and all-organic batteries. The rational design of polymers with tailored functionalities is obscured by the interplay of electronic, ionic, and structural degrees of freedom over a wide range of spatiotemporal scales. We recently developed an effective computational approach that combines physics-based and machine learning techniques to incorporate electronic structure information at large spatiotemporal scales. Leveraging this approach, we investigate the structural, ionic, and electronic properties of mixed ionic-electronic conductors based on redox-active nonconjugated polymers. This emerging class of materials for solid-state organic batteries relies on pendant redox-active sites to transport charge carriers. We systematically examine polymer-electrolyte systems and vary the polymer backbone chemistry, electrolyte solution content, and polymer state of charge. As a function of such system parameters, we then explore in detail structural properties, such as redox active site packing and their interaction with the electrolyte, as well as ionic and electronic transport. The derived relationships between chemical structure, morphology, and ionic and electronic transport inform the design of redox-active polymers with improved characteristics for all-organic battery materials.