Analysis of Charge / Discharge Behavior Using Simple and High-Precision Capacity Measurement: Application to Commercially Available Large-Format Lithium-Ion Batteries

Wednesday, 4 October 2017
Prince George's Exhibit Hall D/E (Gaylord National Resort and Convention Center)
A. Yamazaki (Central Research Institute of electric Power Industry), Y. Kobayashi, Y. Mita, and H. Miyashiro (Central Research Institute of Electric Power Industry)
Lithium-ion batteries are adopted as stationary applications because of their high voltage, high-energy density and high efficiency properties. The batteries for stationary use are expected for long-term operation more than 20 years with 80 % capacity retention. Therefore suitable estimation of the operation period of the lithium-ion batteries are required. High-precision charge / discharge measurement is proposed to estimate an effective life time within short test periods [1]. Then, we developed a simple and high-precision charge / discharge capacity measurement system for large-format cells and determined the capacity fade of 20Ah class battery (SCiBTM by TOSHIBA Corp.). Cells were charged and discharged at 10A(C/2) constant current between 2.7 and 1.5 V at 45(±0.3) °C. The sampling rate of current and voltage during charge / discharge was 100 ms to minimize the counting error of integrated capacity. Figure 1 shows the capacity trends with cycle operation at above condition. The capacity decreased linearly with cycles within the measurement period. The capacity loss per cycle was estimated as 3.4 mAh/cyc by averaging 35 cycles according to the linear approximation as shown in Fig.1. The coulombic loss per cycle was about 8.1 mAh/cyc which was defined as the difference between charge and discharge capacity. This value was more than two times as large as the capacity loss per cycle. The difference between them is probably due to internal reaction without direct involvement of capacity loss during operation. This result suggested that chemical self-discharge process such as shuttle mechanism may be a dominant factor of the estimated coulombic loss.

[1] A. J. Smith, J.C. Burns, S. Trussler, and J. R. Dahn, J. Electrochem. Soc., 157, A196-A202 (2010).