Diagnosis and Prognosis of the Aging of LTO/NMC Li-Ion Cells Under Cycling Tests
Résumé
Novel non-destructive analyses open up possibilities to delve into degradation mechanisms of Lithium-ion batteries and diagnose internal states without the inconvenience of post-mortem characterizations. Low-rate constant current charge or discharge measurements (pOCV(Q)) lower than C/20 give comprehensive information on the electrochemical reactions occurring within the cell. Then, the voltage spectroscopies (Differential Voltage Analysis - dV/dQ vs. Q (DVA) and Incremental Capacity Analysis - dQ/dV vs. V (ICA)) highlight the degradation mechanisms, linked up with the electrode states of health and the available lithium quantity. The aging mechanisms are gathered in four main degradation modes that can be quantified: the loss of lithium inventory (LLI), the loss of positive active materials (LAMPE), the loss of negative active materials (LAMNE) and the ohmic resistance increase (ORI). Estimating the aging laws of the modes enable to forecast the cell useful capacity and predict shifts into the aging trend through the upcoming electrode limitation changes. The diagnostic study is conducted on 28Ah prismatic Li4Ti5O12/Ni1-x-yMnxCoyO2 based cells for different electrical solicitations at 45°C. Cells are cycled between 70% and 100%SOC under 3C or 6C. A 45°C calendar aging test is performed on the mean of the cycling window (85%SOC) to observe the effects of resting at this temperature. No significant capacity fade can be seen in these aging tests, but changes into the cell internal states are expected as the different pOCV(Q) evolve and the resistances increase. Degradation modes are followed through the non-intrusive methods, and then used to make a prognosis on the future evolution of the cell capacity. The present investigation couples the differential methods with half-cell measurements of negative (LTO) and positive (NMC) electrode coin cells. From these data, a full cell model computed from the Alawa toolbox emulates the aging and design degradation maps to spot the focus of interest (FOI) points, i.e. the pOCV(Q) areas where the features of a specific degradation mode can easily be distinguished. The beginning of life cell design imposes that the negative electrode limits the cell operation both at the ends of charge and discharge. Thus, in the first part of the battery’s life, the negative electrode works between insertion rates close to 0 and 1, and its capacity is equivalent to the cell one. In the other hand, a clear FOI of the positive electrode appears on the ICA last shoulder at high voltage. From this knowledge, the positive and negative insertion rates can be described as a function of the cell voltage, and the possible changes into the electrode limitation can be predicted. This approach allows to precisely quantify the degradation modes LLI, LAMPE, LAMNE and ORI for the aging tests carried out. Diagnoses are conducted to find out the aging laws of each mode regarding the different conditions and determine the actual insertion rates that limit the cell operation window. Eventually, this non-intrusive method enables to forecast the next changes of electrode limitation that could affect the aging trend and the cell lifetime.