Testing Lithium Batteries to Destruction
As part of my undergraduate dissertation, my main project was to test lithium batteries for electric vehicle applications.
The project was sponsored by WMG, Jaguar Land Rover and their parent company Tata Motors in conjunction with the School of Engineering at the University of Warwick.
Project Aims and Objectives
The aim of this project was to analyse the thermal stability of lithium ion cells. Therefore, the project objectives were:
- Conduct heat-wait-seek (HWS) and overcharge (OC) abuse testing on lithium ion pouch cells using accelerating rate calorimetry.
- Thermally characterise cells and track the heat output along with the self-heating rate.
- To analyse and compare results to analytical models (where possible) and determine any trends of cell behaviour as a result of the abuse tests.
Calorimetry uses temperature to study or measure properties of materials. The aim was to obtain time-temperature-pressure data of the test samples.
The cell itself was suspended from a copper rod that was free to rotate inside the chamber. Aluminium tape with high grade adhesive was chosen to ensure it would cope with the high temperatures inside the chamber. It was also used to attach the thermocouples to the cell surface. The ARC thermocouple along with one from the Maccor were taped to middle of the front surface. Another Maccor thermocouple was taped to the middle of the back surface, totalling three thermocouples.
A total of 21 experiments where conducted for the dissertation. The data set were analysed and appropriate visuals were created to better understand the failure pathway and draw conclusions.
The results from the overcharge tests spawned another undergraduate project conducted for Jaguar Land Rover for 2016-17.
The general trend from the HWS tests have suggested that the failure time for low SOC cells is shorter compared to high SOC cells. Furthermore, the temperature at which the cell enters thermal runaway increases as the SOC decreases. A proposed theory for this is that at low SOC, the electrodes are more stable. This delays the breakdown of the SEI layer, which is usually the first causality in the breakdown of cells.
This may offer other avenues to safety. For instance, if the BMS of car detects anomalies with the cells, it could quickly discharge the cells to low SOC to reduce the impact of the explosion. In addition, at low SOC the failure temperature rises, which may even prevent the thermal runaway in first instance if the surrounding temperature of the cells is below this threshold.
From the overcharge tests, there is an inverse power relationship between C-rate and the time to failure. As the current increases, the time needed for the cell to fail decreases. The time to failure at C/1000 would be in excess of 88,000 hours. Therefore, it is more likely that the cell would fail as a result of other factors before then.
However, one area of concern for this inverse power relationship could be charging stations. These apply high currents and voltages to the battery pack to reduce the energy transfer time. Therefore, further research regarding fast charging and C-rate may be required to ensure that the technology is safe for consumer use. Appropriate safeguards in both the car and the charging docks may be able to overcome this.