When
Monday, March 16, 2026, at 10:00 a.m.
Vitaliy Yurkiv
Assistant Professor
Aerospace & Mechanical Engineering
University of Arizona
"Mechanisms Governing Battery Thermal Runaway and Early-Warning Prediction: From Interphases to Oxygen Release"
Harshbarger 118A-A1
ABSTRACT: Thermal runaway is a rapid, self-accelerating failure mode in rechargeable batteries that occurs when heat generated by coupled electrochemical and chemical reactions exceeds the cell's ability to dissipate it. As energy density rises and operating windows widen, a key challenge is to predict the onset of instability early enough to enable mitigation, while retaining mechanistic insight into which pathways dominate across chemistry, design and operating conditions. This talk presents an integrated experimental, modeling, and machine-learning framework to understand and predict thermal runaway in rechargeable batteries. Controlled abuse and diagnostic experiments are used to map heat-generation pathways as functions of temperature and state-of-charge, and to connect cell-level signals to the underlying materials-driven reaction sequence. Particular attention is given to precursors that frequently govern the transition to instability, including solid electrolyte interface and cathode electrolyte interface decomposition and their reformation, electrolyte decomposition and gas generation, lithium plating and dendritic growth leading to internal shorting, and, depending on cathode chemistry and state-of-charge, enhanced parasitic reactivity associated with lattice oxygen release and subsequent reactions with electrolyte and electrodes. Various computational methods, such as multi-physics/ab-initio modeling and machine learning, are used to explain experimental measurements of thermal runaway signatures.
BIOSKETCH: Vitaliy R. Yurkivis an assistant professor of aerospace and mechanical Engineering at the University of Arizona. He earned his PhD from Heidelberg University in Germany, then led battery and fuel-cell development efforts at the German Aerospace Center (DLR), Stuttgart, Germany. He subsequently worked as a research professor at the University of Illinois Chicago, before joining the University of Arizona in 2022. Yurkiv has authored over 100 publications, including 67 peer-reviewed papers in leading journals. His recent work integrates multi-physics modeling, high-fidelity experimentation, and machine learning to deliver early-warning diagnostics for lithium-ion battery thermal runaway for electric vehicle and electric airplane applications. This translational focus is reflected in two recent patents. He collaborates closely with Tucson-based companies Ampcera and Sion Power, focusing on the safety of solid-state and Li-metal batteries. He leads a DoD ONR DEPSCoR project on machine-learning–assisted forecasting of thermal events in rechargeable batteries, an NSF CDS&E award on deep-learning analysis of complex materials and an RII Eighteenth Mile TRIF project on AI-driven mitigation of thermal runaway. He also serves as senior personnel and energy-storage lead for the University's Big Idea Challenge project, Making Space for off-Earth Scalable Cloud Computing and Data Infrastructure (AZSCI).