(597g) Multiscale Simulations of Solid-State Redox-Active Polymers for Energy Storage

Radical-containing, redox-active polymers are emerging materials for solid-state energy storage devices. Charge transport in these materials is affected by polymer swelling induced by interactions with the electrolyte, the change of the polymer state of charge during battery functioning, and how these processes reflect on the microstructure-electronic property relationships of a particular material. To gain fundamental understanding on such poorly understood phenomena, there is a need for accurate and efficient multiscale models that incorporate both mesoscale morphological features and electronic structure.

We present results on a multiscale modeling framework that uses physics-based models and machine learning techniques and that allows for a description of the above-mentioned phenomena. First, atomistic molecular dynamics are used to investigate molecular packing as a function of state of charge, swelling, and polymer backbone chemistry. Electronic structure calculations on the sampled conformational spaces then allow to compute electronic couplings and correlate structural and electronic properties. Next, using our recently developed method that combines machine learning and coarse-grained modeling to predict electronic properties at large spatiotemporal scales, we investigate the charge transport properties of these polymers at mesoscopic spatiotemporal scales, providing a link between experiments and simulations. The derived morphology-electronic structure relationships inform the design of radical-containing polymers with improved characteristics for all-organic battery materials.