Abstract
This thesis is based on a recently developed energy storage device called the semi-solid redox flow battery (SSRFB) which represents a technological crossover between two already commercialized systems, namely, lithium ion batteries (LIBs) and redox flow batteries (RFBs). The SSRFB has particle suspensions as electrodes which can be stored outside the discharging battery unit, in tanks, for large scale energy applications. The challenges that plague the SSRFBs are identified in this work and are assigned to the many-body problem where many phenomena compete with one another, across scales leading to a high degree of complexity, thus making experimental results non-intuitive. This work explores the applicability of mechanistic models to handle this type of complexity. The model is an on-lattice, three dimensional, agent based model with a kinetic Monte Carlo engine. Since such models are challenging to validate, the use of a global parameter sensitivity analysis (PSA) is proposed. It is demonstrated that using PSA over large amounts of simulated results can provide trends that help identify emergent mesoscopic metrics to rationalize the complexity of SSRFBs; such insights are not apparent from either experiments or traditional models. The perspectives and tools developed herein can provide a more rigorous framework for other types of electrochemical systems that lie at intersections of disciplines.