Degradation and processes in lithium batteries

Thesis

The automotive industry requires rapid advances in battery technology to fulfil the range and price criteria set by its consumer base. Lithium-ion cell chemistries are currently best-suited to this application, but are approaching their maximum practical energy density and have yet to exhibit lifespans that are equivalent to that of the vehicles that they power. Therefore, there is considerable interest in the next-generation lithium-air cell, which promises upwards of a five-fold increase in energy density over today’s lithium-ion cells. However, there are many hurdles to overcome, the most important being limited cycle life.

Two new organic solvents, namely adiponitrile and glutaronitrile, were investigated as possible new electrolytes for non-aqueous lithium-air cells. Both electrolytes were found to be susceptible to nucleophilic attack by the intermediate O2·- radical species, resulting in rapid capacity loss upon cycling and the conversion of the discharge product, Li2O2, to LiOH. As such, these solvents have been ruled out for use in lithium-air applications.

Work then focussed on the parameterisation of lithium-air cells to assist the development of a multiscale model of the lithium-air discharge mechanism that incorporates the simultaneous formation of Li2O2 in solution and as a surface layer for the first time. Numerous analytical techniques were used to obtain the oxygen concentration and diffusion coefficient of multiple electrolytes, as well as details pertaining to the cathode structure of the cell. The model was shown to predict the discharge profile of a lithium-air cell with greater accuracy than previous models at low and medium current densities, and will thus be a useful tool for the rapid screening of new electrolyte solutions and cathode structures.

Finally, a new 3-electrode modification technique was developed for commercial lithium-ion pouch cells that allows potential profiles of each electrode to be obtained in-situ with no impact upon cell performance, as verified by cycling with unmodified cells and cathode/anode half-cells, and by post-mortem analysis. This technique has provided the data required to experimentally verify a parametric open circuit voltage model for lithium-ion cells that can model and predict electrode-specific decomposition mechanisms to within an accuracy of 10 mV RMSE. Ultimately, the model may be used to improve the effectiveness of a vehicular battery management system and extend the lifespan of the traction battery pack.

Authors: McTurk, E

• DOI: 10.5287/ora-mjn6aw0px
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: Lithium-ion; Lithium-air
• Subjects: Lithium cells; Materials science; Energy storage