(Invited) A Hybrid Capacitor with LFP-Based Positive and AC-Based Negative Electrodes

Abstract

1. Introduction EDLCs have been applied to backup power supplies and auxiliary power sources for memories, because of their feature of high input-output density. It is, however, essential to improve energy density for further application. In recent years, therefore, hybrid capacitors exhibiting higher energy density than that of EDLCs have attracted considerable attention as new electric storage devices to replace EDLCs as well as LIBs.In this research, we have focused on hybrid capacitors utilizing a LFP electrode as a positive electrode, which is used in a LIB, and activated carbon (AC) for a negative electrode. As a point of general concern, it has been reported that a typical solvent for EDLC is reductively decomposed at high voltage operation; the decomposed products are deposited on a negative electrode, and then adsorption sites for cations are lost [1]. In this research, therefore, we aimed at high voltage operation of the hybrid capacitor using 1,2-dimethoxyethane (DME) which is a typical solvent with reduction resistance for an electrolyte.2. Experimental As for the negative electrode material, YP50F, which is micro- and meso-porous activated carbon, was used. AB was used as conductive additive. SBR and CMC were applied to a binder and dispersant, respectively. Regarding the positive electrode, lithium iron phosphate (LFP) was used as active material with ketjen black as conductive additive and PVdF as binder. A two-electrode cell was constructed by using LiTFSI dissolved in DME or PC so as to be 1 mol dm–3 as the electrolyte with a microporous polyolefin membrane as a separator. The N/P ratio was normally 0.694. A galvanostatic cycling test was performed; the maximum operating voltage was 2.5 V.3. Results and Discussion Fig. 1 shows the charge and discharge curves at the first cycle, and Fig. 2 indicates the coulombic efficiency and the capacity retention when the galvanostatic test in 50 cycles is carried out. From the charging curve in Fig. 1, regarding the cell with PC, a large irreversible capacity increase can be confirmed at the charge end when compared to the DME cell. In the discharge curve, the plateau can be seen in PC, but the DME system shows a linear profile. This may be explained by the mechanism that the reduction products of PC, which is generated during charge, is oxidized during discharge. Looking at Fig. 2, the coulombic efficiency at the first cycle is about 54% for PC, while it improves to about 80% for DME.These results suggest that PC undergoes a significant irreversible reaction at the AC negative electrode in the initial charge and discharge. After the third cycle, Both DME and PC cells show a high efficiency of ca. 98%. This means that although PC decomposes considerably until the third cycle and irreversible products continue to be generated, the highly resistive characteristics of the decomposition products would suppress further PC decomposition from the third cycle onwards, and then PC cell also looks stable in terms of coulombic efficiency. The capacity retention of DME is twice as large as that of PC. This may be due to the low contribution of irreversible reaction at the negative electrode at the first cycle in the cell containing DME. From the above results, it was found that when DME is used as a solvent, no decomposition occurs and hence a stable cycle can be obtained.Reference1) S. Ishimoto, Y. Asakawa, M. Shina, and K. Naoi, J. Electrochem. Soc. 156 (7), A563-A571 (2009). Figure 1