Current Compliance‐Driven Morphological Evolution of Conductive Filaments in ECM Cells: Mechanisms, Switching Regimes, and Thermal Effects

Abstract Electrochemical metallization (ECM) cells are a key candidate for emerging applications in neuromorphic computing and in‐memory computing, where precise control over switching dynamics is essential. A major challenge in understanding ECM switching behavior lies in accurately modeling conductive filament evolution, particularly side growth. In this study, the first physical continuum model is presented capable of explaining side growth while incorporating all key mechanisms, including redox reactions at interfaces, ion drift, tunnel currents, mechanical stress, active electrode dissolution, Joule heating, and additional tunnel barrier heating. By accounting for these coupled effects together with a moving mesh algorithm for conductive filament growth, the model captures diverse filament morphologies, which play a crucial role in relaxation times of volatile ECM cells and RESET kinetics in non‐volatile ECM cells. The results provide important insights into filament formation dynamics, with implications for optimizing ECM‐based memory technologies in neuromorphic and in‐memory computing.