Evaluation of HEV Batteries for Recycle ~Investigation for Estimating Battery Capacity By Electrochemical Impedance Spectroscopy~

Abstract

1.Introduction After the end of life of hybrid electric vehicle (HEV), there is a possibility that some of the Nickel-metal hydride (Ni-MH) batteries for HEV remain capacity and we could reuse these batteries without any trouble. In order to utilize these reusable batteries, the capacity estimation method to reuse used batteries is essential. Ni-MH battery capacity decreases during the actual usage. Then we reported that the low capacity retention batteries caused by the negative electrode could be sorted out by the electrochemical impedance spectroscopy (EIS) in a charge state [1.2].On the other hand, it is well known that there are low capacity retention batteries caused by aging of NiOOH in the positive electrode. Among them, the most important thing to consider for reusing the Ni-MH batteries is low capacity retention batteries caused by “memory effect” [3]. When the memory effect occurs, the charging voltage at each SOC increased in accordance with history of the battery. The increase of charging voltage accelerates the side reaction. As a result, capacity decreases due to memory effect. Generally, the memory effect can be eliminated by low rate (0.3C-0.4C) charge and discharge refresh cycles. However, in order to completely eliminate the memory effect, it is necessary that charge and discharge cycles. As a result, it takes a long time to eliminate the memory effect. Thus, in order to determine quickly if the used battery could be reused, it is demanded that the easy method to detect the degree of capacity decrease due to the memory effect. In this study, we would like to report the quick and accurate method to detect the degree of capacity decrease due to the memory effect by EIS. 2.Experimental We used a Ni-MH battery for HEV (1.2 V - 6.5 Ah) as an experimental cell. In order to get the various degrees of capacity decrease due to the memory effect, the cells were charged and discharged for several cycles at constant current of 3 C rate at 35 oC in the range from SOC 0 % to SOC 20 %, SOC 40 %, SOC 60 %, and SOC 80 %. After the accumulated discharge capacity reached 2600 Ah, AC impedance measurements were carried out at SOC 60 %. The AC impedance spectra were obtained with 0.15 C of AC signal in the frequency range of 100 kHz to 0.01 Hz. All measurements were performed at 25 oC. We defined the degree of memory effect as the sum of difference of charging voltage (∫ΔVoltage) between the initial cell and the cycled cell. 3.Results and discussion Fig.1 (a) shows the charge curves of the cells which was obtained on test condition detailed before. We can see the charge curves difference by changing in the range of SOC. Fig.1 (b) shows the AC impedance spectra of these cells. Impedance spectrum of the Ni-MH battery shows semicircle which indicates the charge electron transfer reaction and linear part which indicates the diffusion process. The semicircle in impedance spectrum of the Ni-MH battery is mainly attributed to the negative electrode and the linear part is mainly attributed to the positive electrode. We found that the radius of semicircle in impedance spectrum decreased after cycled which indicate the resistance of electron transfer reaction of negative electrode decreased due to the activation of the negative electrode. On the other hand, in the linear part, we found that the imaginary value on each frequency get higher with the memory effect on the positive electrode. In order to confirm the relationship between the memory effect and impedance spectrum, we calculated (ΔZ"/Δ(ω-1))-1 [AsV-1] as the frequency response of the imaginary components of the linear part. Fig.1(c) shows the relationship between (ΔZ"/Δ(ω-1))-1 and the degree of memory effect (∫ΔVoltage). As shown in Fig.1 (c), (ΔZ"/Δ(ω-1))-1 correlates with the degree of memory effect. Thus, we can estimate the degree of memory effect by calculating (ΔZ"/Δ(ω-1))-1 from the linear part in impedance spectrum, suggesting that it is possible that the degree of capacity decrease due to memory effect can be detected by using (ΔZ"/Δ(ω-1))-1. Reference [1]H. Nishi, D. Koba, S. Ito, T. Yao, D. Mukoyama, H. Nara, S. Tsuda, T. Momma, T. Osaka, The 56rd Battery Symposium in Japan, 1M25 (2015)[2]D. Koba, H. Nishi S. Ito, T. Yao, D. Mukoyama, H. Nara, S. Tsuda, T. Momma, T. Osaka, The 56rd Battery Symposium in Japan, 1M26 (2015) [3] Y. Sato, S. Takeuchi, K.Kobayakawa, J. Power Sources 93 (2001) 20-24 Figure 1