Mapping safety transitions as batteries degrade: A model-based analysis towards full-lifespan battery safety management

Xinlei Gao,Ruihe Li,Gregory J. Offer,Huizhi Wang

2024-08-29

Abstract:Battery safety is important, yet safety limits are normally static and do not evolve as batteries degrade. Consequently, many battery systems are overengineered to meet increasingly stringent safety demands. In this work we show that failure behaviour evolves over time as batteries degrade, and discuss the challenges and opportunities to manage battery safety dynamically throughout its lifetime. We introduce the first framework for capturing how the likelihood and severity of battery failures change over time based upon the concepts of safety zones and their boundaries. Through the development of a comprehensive physics-based model that integrates multiple degradation and thermal runaway failure mechanisms, we then show how the safety zones and boundaries of a commercial 21700 battery change after varied use and how these changes may lead to false negatives with existing management strategies. Further analyses reveal that degradation mechanisms strongly affect safety characteristics, causing significant changes despite similar capacity fade, highlighting the limitations of using capacity fade alone to assess batteries' usability. By synthesising our results with literature, we map possible degradation-to-failure pathways and recommend future research needs to achieve full-lifespan battery safety management, with advanced diagnostic and modelling techniques to accurately define state-of-safety for real-world applications as key priorities.

Chemical Physics

What problem does this paper attempt to address?

The paper aims to address the issue of static safety limits in battery safety management systems and explores the dynamic changes in safety characteristics due to degradation over the battery's lifespan. Specifically:

1. **Existing Issues**: Current Battery Management Systems (BMS) typically use static safety limits to assess battery reliability. These limits are set relatively conservatively, yet misjudgments still occur. This leads to many battery systems being over-engineered to meet increasingly stringent safety requirements, sacrificing performance and cost-effectiveness.

2. **Research Objectives**: The paper proposes a new framework to capture the changes in battery failure behavior over time and discusses the challenges and opportunities of dynamically managing battery safety throughout its lifecycle. By establishing a comprehensive physical model that integrates multiple degradation mechanisms and thermal runaway failure mechanisms, the study demonstrates how the safety regions and boundaries of commercial 21700 batteries change under different usage conditions and analyzes how these changes may lead to misjudgments in existing management strategies.

3. **Core Contributions**: Through the study of multiple degradation paths, the paper reveals that different usage patterns can lead to significant differences in battery safety characteristics, even if the capacity fade is similar. It emphasizes that solely relying on capacity fade to assess battery safety is insufficient and suggests future research directions to achieve full lifecycle battery safety management.

In summary, the paper attempts to address the issue of dynamic changes in battery safety limits over time and provides new methods for understanding and predicting these dynamic changes.