Advanced Diagnostics for Lithium-ion Batteries: Decoding the Information in Electrode Swelling

Abstract

Lithium-ion batteries exhibit mechanical expansion and contraction during cycling, consisting of a reversible intercalation-induced expansion and an irreversible expansion as the battery ages. Prior experimental studies have shown that mechanical expansion contains valuable information that correlates strongly with cell aging. However, a number of fundamental questions remain on the usability of the expansion measurement in practice. For example, it is necessary to determine whether the expansion measurements provide information that can help the estimation of the electrode state of health (eSOH), given limits on data availability and sensor noise in the field. Furthermore, the viability of using expansion for cell diagnostics also needs more investigation considering the broad range of aging conditions in real-world applications.

This dissertation focuses on the experimental and modeling study of the expansion measurements during aging in order to assess its ability in helping battery diagnostics. To this end, mechanistic voltage and expansion models based on the underlying physics of phase transitions are developed. For the first time, the identifiability of eSOH parameters is explored by incorporating the expansion/force measurement. It is shown that the expansion measurements enhance the estimation of eSOH parameters, especially with a limited data window, since it has a better signal-to-noise ratio compared to the voltage. Moreover, the increased identifiability is closely related to the phase transitions in the electrodes.

A long-term experimental aging study of the expansion of the graphite/NMC pouch cells is conducted under a variety of conditions such as temperature, charging rate, and depth of discharge. The goals here are to validate the results of the identifiability analysis and record the reversible and irreversible expansion correlated with capacity loss for informing the electrochemical models. Firstly, the advantages of the expansion concerning the eSOH identifiability are confirmed. Secondly, the results of the expansion evolution reveal that the features in the reversible expansion are an excellent indicator of health and, specifically, capacity retention. The expansion feature is robust to charge conditions. Namely, it is mostly insensitive to the hysteresis effects of the various initial state of charge, and it is detectable at higher C-rates up to 1C. Additionally, the expansion feature occurs near the half-charged point and therefore diagnostics can be performed more often during naturalistic use cases. Thus, the expansion measurement facilitates more frequent capacity checks in the field.

Furthermore, an electrochemical and expansion model suitable for model-based estimation is developed. In particular, a multi-particle modeling approach for the graphite electrode is considered. It is demonstrated that the new model is able to capture the peak smoothing effect observed in the differential voltage at higher C-rates above C/2. Model parameters are identified using experimental data from the graphite/NMC pouch cell. The proposed model produces an excellent fit for the observed electric and mechanical swelling response of the cells and could enable physics-based data-driven degradation studies at practical charging rates.

Finally, a fast-charging method based on the constant current constant voltage (CC-CV) charging scheme, called CC-CVησT (VEST), is developed. The new approach is simpler to implement and can be used with any model to impose varying levels of constraints on variables pertinent to degradation, such as plating potential and mechanical stress. The capabilities of the new CC-CVησT charging are demonstrated using the physics-based model developed in this dissertation.