RESEARCH
Battery Safety
Battery safety will become increasingly crucial as the battery market grows along with the demand for transportation electrification and the increasing energy density in battery systems. Reliable fault detection (leaks, venting gas, temp, pres, and voltage) and effective mitigation methods (discharge & cooling) are required to alert, advise on needed actions based on predicting evolution, and reduce the risk. Our group is developing methods to address these issues and guide emergency response to better prevent and eliminate the hazards.
Publications
Ting Cai, et al Detection of Li-ion battery failure and venting with Carbon Dioxide sensors. eTransportation, 2021. [ bib | DOI ]
Ting Cai, et al. Modeling li-ion battery temperature and expansion force during the early stages of thermal runaway triggered by internal shorts. JECS (2019) [ bib | DOI ]
T. Cai, V. Tran, et al, "Modeling Li-ion Battery First Venting Events Before Thermal Runaway", Modeling, Estimation, and Control Conference (MECC), 2021 [pre-print]
Battery reversible expansion corresponds to electrode material phase transitions. Unlike the differential voltage, the double differential expansion peaks are observable up to 1C rate, at low depth of discharge, and are less path-dependent, which makes the differential expansion an excellent method for capacity and health estimation. To understand why that is the case, and at the same time develop a model-based estimator a reduced ordered electrochemical/mechanical model is developed. The electrochemical model is primarily based on the well-known Doyle Fuller Newman (DFN) model and includes a particle expansion model.
Publications
Mohtat, P., et al (2022), “Comparison of expansion and voltage differential indicators for battery capacity fade,” JPS [DOI]
Mohtat, P., et al (2021). “Reversible and Irreversible Expansion of Lithium-Ion Batteries Under a Wide Range of Stress Factors”. JECS. [DOI]
Mohtat, P., et al. (2021). "An Algorithmic Safety VEST For Li-ion Batteries During Fast Charging." Modeling, Estimation and Controls Conference [preprint].
Mohtat, P., at al. (2020). “Differential Expansion and Voltage Model for Li-ion Batteries at Practical Charging Rates”. JECS, 167(11), 110561. [preprint | DOI]
Battery Degradation Modeling and Lifetime Prediction
To enable reusability of batteries (2nd Life Use) there is a need for enhanced modeling for prediction and accelerated simulation of how batteries degrade.
We model physics-based degradation mechanisms (SEI growth, lithium plating, and particle swelling and cracking) to understand their interactions and predict the knee point where accelerated aging starts. Classical and model-predictive control techniques can then guide degradation-conscious EV fast-charging, 2nd life BMS, adaptive power denials, ESS derating and allow adaptive accelerated simulations of a battery's entire life.
Finally, we also work with data-driven methods to predict battery lifetime based on features in the charge/discharge curves.
Publications
Pannala, et al. (2021) "Methodology for Accelerated Inter-Cycle Simulations of Li-ion Battery Degradation with Intra-Cycle Resolved Degradation Mechanisms". Submitted to American Control Conference 2022.
Sulzer, V., et al. (2020). “Promise and Challenges of a Data-Driven Approach for Battery Lifetime Prognostics”. Submitted to American Control Conference 2021. [preprint]
Manufacturing Diagnostics
The ability to quickly assess battery lifetime is critical for enabling manufacturers to make smart design choices that balance initial performance with longevity.
Toward this goal, we are analyzing data generated during the battery manufacturing process to identify diagnostic signals that can be used to screen new materials and manufacturing processes rapidly.
Partnering with the University of Michigan Battery Lab, we are building prototype pouch cells with varying process parameters and deploying low-cost signals into manufacturing. We also perform cycle life characterization to study the connection between early-life signals and end-of-life outcomes.
Publications
A. Weng, E. Olide, I. Kovalchuk, J.B. Siegel, A. Stefanopoulou, "Modeling Battery Formation: Boosted SEI Growth, Multi-Species Reactions, and Irreversible Expansion." Journal of the Electrochemical Society 2023, 170 (9), 090523
A. Weng, J.B. Siegel, A. Stefanopoulou, "Differential voltage analysis for battery manufacturing process control." Frontiers in Energy Research 2023, 11, 1087269
A. Weng, P. Mohtat, P. M. Attia, V. Sulzer, S. Lee, A. Stefanopoulou, "Predicting the impact of formation protocols on battery lifetime immediately after manufacturing" Joule 5, 1-22, November 17, 2021, doi: 10.1016/j.joule.2021.09.015.
EV Manufacturing Workforce Trends
Figure from our paper "30% fewer workers for electric vehicle assembly": harbinger or myth? by Andrew Weng, Omar Ahmed, Gabriel Ehrlich, and Anna Stefanopoulou, currently under review at Nature Communications (pre-print here).
It has been widely hypothesized that the EV transition will require 30% fewer assembly workers compared to internal combustion engine vehicles. To investigate this claim, we studied the U.S. auto assembly workforce landscape (fig. a) and for the first time examined what we call "Transition Plants" (fig. b): assembly plants that have previously assembled internal combustion vehicles but have since fully transitioned to assembling battery electric vehicles.
We used publicly available datasets on vehicle production and employment to show that three Transition Plants in the U.S. have all required more, not fewer, workers to assemble the same number of vehicles after transitioning to making EVs.
Our work, currently under review at Nature Communications, suggests that widespread loss of employment at electric vehicle assembly sites is a smaller risk than many fear. Moreover, our study serves as a call for more regionally-focused analyses of the EV transition's effects on labor using data-driven and macro-level surveying approaches.